Skip to main content

Full text of "The Sarva-Darsana-Samgraha, or Review of the different systems of Hindu philosophy"

See other formats


" A knowledge of the commonplace, at least, of Oriental literature, philo 
sophy, and religion is as necessary to the general reader of the present day 
as an acquaintance with the Latin and Greek classics was a generation or so 
ago. Immense strides have been made within the present century in these 
branches of learning; Sanskrit has been brought within the range of accurate 
philology, and its invaluable ancient literature thoroughly investigated ; the 
language and sacred books of the Zoroastrians have been laid bare ; Egyptian, 
Assyrian, and other records of the remote past have been deciphered, and a 
group of scholars speak of still more recondite Accadian and Hittite monu 
ments ; but the results of all the scholarship that has been devoted to these 
subjects have been almost inaccessible to the public because they were con 
tained for the most part in learned or expensive works, or scattered through 
out the numbers of scientific periodicals. Messrs. TRUBNER & Co., in a spirit 
of enterprise which does them infinite credit, have determined to supply the 
constantly-increasing want, and to give in a popular, or, at least, a compre 
hensive form, all this mass of knowledge to the world." Times. 

Second Edition, post 8vo, pp. xxxii. 748, with Map, cloth, price 213. 


By the HON. SIR W. W. HUNTER, K.C.S.I., C.S.I., C.I.E., LL.D., 

Member of the Viceroy s Legislative Council, 
Director-General of Statistics to the Government of India. 

Being a Revised Edition, brought up to date, and incorporating the general 
results of the Census of 1881. 

" It forms a volume of more than 700 pages, and is a marvellous combination of 
literary condensation and research. It gives a complete account of the Indian 
Empire its history, peoples, and products, and forms the worthy outcome 01 
seventeen years of labour with exceptional opportunities for rendering that labour 
fruitful Nothing could be more lucid than Sir William Hunter s expositions of the 
economic and political condition of India at the present time, or more interesting 
than his scholarly history of the India of the past. The limes. 


Third Edition, post 8vo, cloth, pp. xvi. 428, price i6s. 




Late of the Universities of Tiibingen, Gottingen, and Bonn ; Superintendent 
o.f Sanskrit Studies, and Professor of Sanskrit in the Poona College. 

To which is added a Biographical Memoir of the late Dr. HAUG 

by Prof. E. P. EVANS. 
I. History of the Researches into the Sacred Writings and Religion of the 

Parsis, from the Earliest Times down to the Present. 
II. Languages of the Parsi Scriptures. 

III. The Zend-Avesta, or the Scripture of the Parsis. 

IV. The Religion, as to its Origin and Development. 

" Essays on the Sacred Language, Writings, and Religion of the Parsis, by the 
late Dr. Martin Hang, edited by Dr. E. W. West. The author intended, on his return 
from India, to expand the materials contained in this work into a comprehensive 
account of the Zoronstriaii religion, but the design was frustrated by his untimely 
death. We have, however, in a concise and readable form, a history of the researches 
into the sacred writings and religion of the Parsis from the earliest times down to 
the present a dissertation on the languages of the Parsi Scriptures, a translation 
of the Zend-Avesta, or the Scripture of the Parsis, and a dissertation on the Zoroas- 
trian religion, with especial reference to its origin and development. "Times. 

Post 8vo, cloth, pp. viii. 176, price 75. 6d. 



With Accompanying Narratives. 

Translated from the Chinese by S. BEAL, B.A., Professor of Chinese, 
University College, London. 

The Dhammapada, as hitherto known by the Pali Text Edition, as edited 
by Fausboll, by Max Miiller s English, and Albrecht Weber s German 
translations, consists only of twenty-six chapters or sections, whilst the 
Chinese version, or rather recension, as now translated by Mr. Beal, con 
sists of thirty-nine sections. The students of Pali who possess Fausboll s 
text, or either of the above named translations, will therefore needs want 
Mr. Beal s English rendering of the Chinese version ; the thirteen above- 
named additional sections not being accessible to them in any other form ; 
for, even if they understand Chinese, the Chinese original would be un 
obtainable by them. 

"Mr. Beal s rendering of the Chinese translation is a most valuable aid to the 
critical study of the work. It contains authentic texts gathered from ancient 
canonical books, and generally connected with some incident in the history of 
Buddha. Their great interest, however, consists in the light which they throw upon 
everyday life in India at the remote period at which they were written, and upon 
the method of teaching adopted by the founder of the religion. The method 
employed was principally parable, and the simplicity of the tales and the excellence 
of the morals inculcated, as well as the strange hold which they have retained upon 
the minds of millions of people, make them a very remarkable study." Times. 

" Mr. Beal, by making it accessible in an English dress, has added to the great ser 
vices he has already rendered to the comparative study of religious history." Academy. 

"Valuable as exhibiting the doctrine of the Buddhists in its purest, least adul 
terated form, it brings the modern reader face to face with that simple creed and rule 
of conduct which won its way over the minds of myriads, and which is now nominally 
professed by 145 millions, who have overlaid its austere simplicity with innumerable 
ceremonies, forgotten its maxims, perverted its teaching, and so inverted its leading 
principle that a religion whose founder denied a God, now worships that founder as 
a god himself. Scotsman. 


Second Edition, post 8vo, cloth, pp. xxiv. 360, price IDS. 6d. 



Translated from the Second German Edition by JOHN MANN, M.A., and 
THEODOR ZACHARIAE, Ph.D., with the sanction of the Author. 

Dr. BUHLER. Inspector of Schools in India, writes: "When I was Pro 
fessor of Oriental Languages in Elphinstone College, I frequently felt the 
want of such a work to which I could refer the students." 

Professor COWELL, of Cambridge, writes : "It will be especially useful 
to the students in our Indian colleges and universities. I used to long for 
such a book when I was teaching in Calcutta. Hindu students are intensely 
interested in the history of Sanskrit literature, and this volume will supply 
them with all they want on the subject." 

Professor WHITNEY, Yale College, Newhaven, Conn., U.S.A., writes : 
" I was one of the class to whom the work was originally given in the form 
of academic lectures. At their first appearance they were by far the most 
learned and able treatment of their subject ; and with their recent additions 
they still maintain decidedl} 7 the same rank." 

" Is perhaps the most comprehensive and lucid survey of Sanskrit literature 
extant. The essays contained in the volume were originally delivered as academic 
lectures, and at the time of their first publication were acknowledged to be by far 
the most learned and able treatment of the subject. They have now been bnnight 
up to date by the addition of all the most important results of recent research." 

Post 8vo, cloth, pp. xii. 198, accompanied by Two Language 
Maps, price ys. 6d. 



The Author has attempted to fill up a vacuum, the inconvenience of 
which pressed itself on his notice. Much had been written about the 
languages of the East Indies, but the extent of our present knowledge had 
not even been brought to a focus. It occurred to him that it might be of 
use to others to publish in an arranged form the notes which he had collected 
for his own edification. 

" Supplies a deficiency which has long been felt." Times. 

" The book before us is then a valuable contribution to philological science. It 
passes under review a vast number of languages, and it gives, or professes to give, in 
every case the sum and substance of the opinions and judgments of the best-informed 
writers. " Saturday Review. 

Second Corrected Edition, post 8vo, pp. xii. 116, cloth, price 53. 



Translated from the Sanskrit into English Verse by 

" A very spirited rendering of the Kumdrasambhava, which was first published 
twenty-six years ago, and which we are glad to see made once more accessible." 

"Mr. Griffith s very spirited rendering is well known to most who are at all 
interested in Indian literature, or enjoy the tenderness of feeling and rich creative 
imagination of its author." Indian Antiquary. 

" We are very glad to welcome a second edition of Professor Griffith s admirable 
translation. Few translations deserve a second edition better." Athenwum. 


Post 8vo, pp. 432, cloth, price i6s. 




Late Professor of Hindustani, Staff College. 

"This not only forms an indispensable book of reference to students of Indian 
literature, but is also of great general interest, as it gives in a concise and easily 
accessible form all that need be known about the personages of Hindu mythology 
whose names are so familiar, but of whom so little is known outside the limited 
circle of savants." Times. 

" It is no slight gain when such subjects are treated fairly and fully in a moderate 
space ; and we need only add that the few wants which we may hope to see supplied 
in new editions detract but little from the general excellence of Air. Dowson s work." 
Saturday Review. 

Post 8vo, with View of Mecca, pp. cxii. 172, cloth, price gs. 



Translator of " The Thousand and One Nights ; " &c., &c. 
A New Edition, Revised and Enlarged, with an Introduction by 


"... Has been long esteemed in this country as the compilation of one of the 
greatest Arabic scholars of the time, the late Mr. Lane, the well-known translator of 
the Arabian Nights. . . . The present editor has enhanced the value of his 
relative s work by divesting the text of a great deal of extraneous matter introduced 
by way of comment, and prefixing an introduction." Times. 

" Mr. Poole is both a generous and a learned biographer. . . . Mr. Poole tells us 
the facts ... so far as it is possible for industry and criticism to ascertain them, 
and for literary skill to present them in a condensed and readable form." English 
man, Calcutta. 

Post 8vo, pp. vi. 368, cloth, price 143. 



Hon. LL.D. of the University of Calcutta, Hon. Member of the Bombay Asiatic 

Society, Bodeii Professor of Sanskrit in the University of Oxford. 
Third Edition, revised and augmented by considerable Additions, 

with Illustrations and a Map. 

" In this volume we have the thoughtful impressions of a thoughtful man on some 
of the most important questions connected with our Indian Empire. . . . An en 
lightened observant man, travelling among an enlightened observant people, Professor 
Monier Williams has brought before the public in a pleasant form more of the manners 
and customs of the Queen s Indian subjects than we ever remember to have seen in 
uuy one work. He not only deserves the thanks of every Englishman for this able 
contribution to the study of Modern India a subject with which we should be 
specially familiar but he deserves the thanks of every Indian, Parsee or Hindu, 
Buddhist and Moslem, for his clear exposition of their manners, their creeds, and 
their necessities." Times. 

Post 8vo, pp. xliv. 376. cloth, price 148. 


With an Introduction, many Prose Versions, and Parallel Passages from 

Classical Authors. 
BY J. MUIR, C.I.E., D.C.L., LL.D., Ph.D. 

" . . , An agreeable introduction to Hindu poetry." Times. 

"... A volume which may be taken as a fair illustration alike of the religious 
and moral sentiments and of the legendary lore of the best Sanskrit writers." 
Edinburah Daily Review. 


Second Edition, post 8vo, pp. xxvi. 244, cloth, price IDS. 6d. 



Translated for the First Time into Prose and Verse, with an Introductory 

Preface, and a Life of the Author, from the Atish Kadah, 
" It is a very fair rendering of the original." Times. 

" The new edition has long been desired, and will be welcomed by all who take 
any interest in Oriental poetry. The Gulistan is a typical Persian verse-book of the 
highest order. Mr. Eastwick s rhymed translation . . . has long established itself m 
u secure position as the best version of Sadi s finest work." Academy. 
" It is both faithfully and gracefully executed." Tablet. 

In Two Volumes, post 8vo, pp. viii. 408 and viii. 348, cloth, price 28s. 




Late of the Bengal Civil Service ; Corresponding Member of the Institute; Chevalier 

of the Legion of Honour ; late British Minister at the Court of Nepal, &c., &c. 


SECTION I. On the Kocch, Bodo, and Dhimal Tribes.-Part L Vocabulary. 
Part II Grammar. Part III. Their Origin, Location, Numbers, Creed, Customs, 
Character, and Condition, with a General Description of the Climate they dwell in 

SECTION IL-On Himalayan Ethnology.-I. Comparative Vocabulary of the Lan 
guages of the Broken Tribes of ^paL-H. Vocabulary of the Dia ects of the kmiiti 
Lai^iuige.-III. Grammatical Analysis of the Vayu Language. The \ayu Giammai. 
-IV Analysis of the Billing Dialect of the Kiranti Language. Ihe Lahing Gram- 
111!ir _V On the Vayu or Hayu Tribe of the Central Hin.alaya.-VI. On tne Kiranti 
Tribe of the Central Himalaya. 


SECTION III. On the Aborigines of North-Eastern India. Comparative Vocabulary 
of the Tibetan, Bodo, and Garo Tongues. 

SECTION IV. Aborigines of the North-Eastern Frontier. 

SECTION V. Aborigines of the Eastern Frontier. 

SECTION VI The Indo-Chinese Borderers, and their connection with the . Hima- 
1-tyans and Tibetans. Comparative Vocabulary of Indo-Chinese Borderers in Arakan. 
Comparative Vocabulary of Indo-Chinese Borderers in Teuasserim. 

SECTION VII. The Mongolian Affinities of the Caucasians. Comparison and Ana 
lysis of Caucasian and Mongolian. Words. 

SECTION VIII. Physical Type of Tibetans. 

SKCTION IX. -The Aborigines of Central India.-Comparative Vocabulary of the 
Aborteinal L-uijmacres of Central India. Aborigines of the Eastern Ghats.-Vocabu- 
llr of sonJ "of ? &Malecte < of the Hill and Wandering Tribes in ^^J Q 
-Aborigines of the Nilgiris, with Remarks on their Affimties.-S pplemen. to the 
Nilgirian Vocabularies. -The Aborigines of Southern India and Ceylon. 

SECTION X. -Route of Mission to Pekiri, with Remarks on the Water- 
Shed and Plateau of Tibet. 

SECTION XL-Route from Kathmandu, the Capital of Nepal, to Darjeehng m 
Sikim. Memorandum, relative to the Seven Cosis of JSepai. _ 

SECTION Xll.-Some Accounts of the Systems of Law and Police as recogms 
the State of Nepal. 

SECTION XIII.-The Native Method of making the Paper denominated 1 ta,n, 

^ SECTION XIV -Pre-eminence of the Vernaculars; or, the Anglicists Answered ; 
Being Letters on the Education of the People of India. 

For the study of the less-known races of India Mr. Brian Hodgson .^ ; M*f c "?; 
ous Essays will be found very valuable both to the philologist and the e 


Third Edition, Two Vols., post 8vo, pp. viii. 268 and viii. 326, cloth, 
price 2is. 


The Ways to Neibban, and Notice on the Phongyies or Burmese Monks. 

Bishop of Ramatha, Vicar- Apostolic of Ava and Pegu. 

"The work is furnished with copious notes, wnich not only illustrate the subject- 
matter, but form a perfect encyclopedia of Buddhist lore." Times. 

" A work which will furnish European students of Buddhism with a most valuable 
help in the prosecution of their investigations." Edinburgh Daily Review. 

" Bishop Bigandet s invaluable work. " Indian Antiquary. 

"Viewed in this light, its importance is sufficient to place students of the subject 
under a deep obligation to its author." Calcutta Review. 

" This work is one of the greatest authorities upon Buddhism." Dublin Review. 

Post 8vo, pp. xxiv. 420, cloth, price i8s. 


Author of " China s Place in Philology," "Religion in China," &c., &c. 

"It contains a vast deal of important information on the subject, such as is only 
to be gained by long-continued study on the spot."-^At/ienceum. 

" Upon the whole, we know of no work comparable to it for the extent of its 
original research, and the simplicity with which this complicated system of philo 
sophy, religion, literature, and ritual is set forth." British Quarterly Review. 

"The whole volume is replete with learning. ... It deserves most careful study 
from all interested in the history of the religions of the world, and expressly of those 
who are concerned in the propagation of Christianity. Dr. Edkins notices in terms 
of just condemnation the exaggerated praise bestowed upon Buddhism bv recent 
English writers. "Record. 

Post 8vo, pp. 496, cloth, price ics. 6d. 



Late Member of Her Majesty s Indian Civil Service ; Hon. Secretary to 

the Royal Asiatic Society; 
and Author of " The Modern Languages of the East Indies." 

" We know none who lias described Indian life, especially the life of the natives 
with so much learning, sympathy, and literary talent." Academy. 

" They seem to us to be full of suggestive and original remarks." St. James s Gazette. 

11 His book contains a vast amount of information. The result of thirty-five years 
of inquiry, reflection, and speculation, and that on subjects as full of fascination as 
of food for thought." Tablet. 

" Exhibit such a thorough acquaintance with the history and antiquities of India 
as to entitle him to speak as one having authority. " Edinburgh Daily Review. 

" The author speaks with the authority of personal experience It is this 

constant association with the country and the people which gives such a vividness 
to many of the pages." Athenceum. 


Post 8vo, pp. civ. 348, clotli, price i8s. 


The Oldest Collection of Folk-lore Extant : 


For the first time Edited in the original Pali. 


And Translated by T. W. RHYS DAVIDS. 

Translation. Volume I. 

"These are tales supposed to have been told by the Buddha of what he had seen 
and heard in his previous births. They are probably the nearest representatives 
of the original Aryan stories from which sprang the folk-lore of Europe as well as 
India. Tiie introduction contains a most interesting disquisition on the migrations 
of these fables, tracing their reappearance in the various groups of folk-lore legends. 
Among other old friends, we meet with a version of the Judgment of Solomon." Times. 

" It is now some years since Mr. Rhys Davids asserted his right to be heard on 
this subject by his able article 011 Buddhism in the new edition of the Encyclopaedia 
Britannica. " Leeds Mercury. 

" All who are interested in Buddhist literature ought to feel deeply indebted to 
Mr. Rhys Davids. His well-established reputation as a Pali scholar is a sufficient 
guarantee for the fidelity of his version, and the style of his translations is deserving 
of high praise." Academy. 

" No more competent expositor of Buddhism could be found than Mr. Rhys Davids. 
In the Jataka book we have, then, a priceless record of the earliest imaginative 
literature of our race ; and ... it presents to us a nearly complete picture of the 
social life and customs and popular beliefs of the common people ot Aryan tribes, 
closely related to ourselves, just as they were passing through the first stages of 
civilisation." St. James s Gazette. 

Post 8vo, pp. xxviii. 362, cloth, price 148. 



Compiled and Translated by PAUL ISAAC HERSHON, 

Author of " Genesis According to the Talmud," &c. 

With Notes and Copious Indexes. 

" To obtain in so concise and handy a form as this volume a general idea of the 
Talmud is a boon to Christians at least." Times. 

" Its peculiar and popular character will make it attractive to general readers. 
Mr. Hershon is a very competent scholar. . . . Contains samples of the good, bad, 
and indifferent, and especially extracts that throw light upon the Scriptures." 
British Quarterly Review. 

" Will convey to English readers a more complete and truthful notion of the 
Talmud than any other work that has yet appeared." Daily News. 

"Without overlooking in the slightest the several attractions of the previous 
volumes of the Oriental Series. we have no hesitation in saying that this surpasses 
them all in interest." Edinburgh Daily Review. 

" Mr. Hershon has . . . thus given English readers what is, we believe, a fair set 
of specimens which they can test for themselves." The Record. 

" This book is by far the best fitted in the present state of knowledge to enable the 
general reader to gain a fair and unbiassed conception of the multifarious contents 
of the wonderful miscellany which can only be truly understood so Jewish pride 
asserts by the life-long devotion of scholars of the Chosen People." Inquirer. 

" The value and importance of this volume consist in the fact that scarcely a singJc 
extract is given in its pages but throws some light, direct or refracted, upon those 
Scriptures which are the common heritage of Jew and Christian alike. John Bull. 

" It is a capital specimen of Hebrew scholarship ; a monument of learned, loving, 
light-giving labour." Jewish Herald. , 


Post 8vo, pp. xii. 228, cloth, price 73. 6d. 



Author of " Yeigo Henkaku Shiran." 

" A very curious volume. The author has manifestly devoted much labour to the 
task of studying the poetical literature of the Japanese, and rendering characteristic 
specimens into English verse." Daily News. 

" Mr. Chamberlain s volume is, so far as we are aware, the first attempt which has 
been made to interpret the literature of the Japanese to the Western world. It is to 
the classical poetry of Old Japan that we must turn for indigenous Japanese thought, 
and in the volume before us we have a selection from that poetry rendered into 
graceful English verse." Tablet. 

" It is undoubtedly one of the best translations of lyric literature which has 
appeared during the close of the last year." Celestial Empire. 

"Mr. Chamberlain set himself a difficult task when he undertook to reproduce 
Japanese poetry in an English form. But he has evidently laboured con amore, and 
his efforts are successful to a degree." London and China Express. 

Post 8vo, pp. xii. 164, cloth, price 103. 6d. 

THE HISTORY OF ESARHADDON (Son of Sennacherib), 


Translated from the Cuneiform Inscriptions upon Cylinders and Tablets in 
the British Museum Collection ; together with a Grammatical Analysis 
of each Word, Explanations of the Ideographs by Extracts from the 
Bi-Lingual Syllabaries, and List of Eponyms, &c. 


Assyrian Exhibitioner, Christ s College, Cambridge. 

"Students of scriptural archaeology will also appreciate the History of Esar- 
hadclon. " Times. 

" There is much to attract the scholar in this volume. It does not pretend to 
popularise studies which are yet in their infancy. Its primary object is to translate, 
but it does not assume to be more than tentative, and it offers both to the professed 
Assyriologist and to the ordinary non-Assyriological Semitic scholar the means of 
controlling its results." Academy. 

"Mr. Budge s book is, of course, mainly addressed to Assyrian scholars and 
students. They are not, it is to be feared, a very numerous class. But the more 
thanks are due to him on that account for the way in which he has acquitted himself 
in his laborious task." Tablet. 

Post 8vo, pp. 448, cloth, price 215. 



Book the First. 
Together ^v^th some Account of the Life and Acts of the Author, 

of his Ancestors, and of his Descendants. 
Illustrated by a Selection of Characteristic Anecdotes, as Collected 

by their Historian, 

Translated, and the Poetry Versified, in English, 

" A complete treasury of occult Oriental lore." Saturday Revieic. 

"This book a very valuable help to the reader ignorant of Persia, who is 
desirous of obtaining an insight into a very important department of the literature 
extant in that language." Tablet. 


Post 8vo, pp. xvi. 280, cloth, price 6s. 



Member of the Bengal Asiatic Society, F.R.G.S. 

" We regard the book as valuable, and wish for it a wide circulation and attentive 
reading. " Record. 

" Altogether, it is quite a feast of good things." Globe. 
" It is full of interesting matter." Antiquary. 

Post 8vo, pp. viii. 270, cloth, price ys. 6d. 

Containing a New Edition of the "Indian Song of Songs," from the Sanscrit 
of the "Gita Govinda" of Jayadeva ; Two Books from "The Iliad of 
India" (Mahabharata), " Proverbial Wisdom " from the Shlokas of the 
Hitopadesa, and other Oriental Poems. 
BY EDWIN ARNOLD, C.S.I., Author of "The Light of Asia." 

" In this new volume of Messrs. Triibiier s Oriental Series, Mr. Edwin Arnold does 
good service by illustrating, through the medium of his musical English melodies, 
the power of Indian poetry to stir European emotions. The Indian fSong of Songs 
is not unknown to scholars. Mr. Arnold will have introduced it among popular 
English poems. Nothing could be more graceful and delicate than the shades by 
which Krishna is portrayed in the gradual process of being weaned by the love of 

Beautiful Radha, jasmine-bosomed Hadha, 
from the allurements of the forest nymphs, in whom the five senses are typified. "- 

"No other English poet has ever thrown his genius and his art so thoroughly into 
the work of translating Eastern ideas as Mr. Arnold has done in his splendid para 
phrases of language contained in these mighty epics." Daily Telegraph. 

" The poem abounds with imagery of Eastern luxuriousness and sensuousn. ss ; the 
air seems laden with the spicy odours of the tropics, and the verse has a richness and 
a melody sufficient to captivate the senses of the dullest."- Standard. 

" The translator, while producing a very enjoyable poem, has adhered with toler 
able fidelity to the original text." Overland Mail. 

"We certainly wish Mr. Arnold success in his attempt to popularise Indian 
classics, that being, as his preface tells us, the goal towards which he bends his 
efforts," Allen s Indian Mail. 

Post 8vo, pp. xvi. 296, cloth, price ics. 6d. 




Translated from the Original Text and Classified, with 

Comments and Explanations, 
By the REV. ERNST FABER, Rhenish Mission Society. 

Translated from the German, with Additional Notes, 

By the REV. A. B. HUTCHINSON, C.M.S., Church Mission, Hong Kong. 
"Mr Faber is already well known in the field of Chinese studies by his digest of 
the doctrines of Confucius. The value of this work will be perceived when it is 
remembered that at no time since relations commenced between China and t 
West has the former been so powerful-we had almost said aggressive-as now 
For those who will give it careful study, Mr. Faber s work is one of the most 
valuable of the excellent series to which it belongs." Nature. 

A 2 


Post 8vo, pp. 336, cloth, price i6s. 


Translated from the French with the authority and assistance of the Author. 

The author has, at the request of the publishers, considerably enlarged 
the work for the translator, and has added the literature of the subject to 
date ; the translation may, therefore, be looked upon as an equivalent of a 
new and improved edition of the original. 

" Is not only a valuable manual of the religions of India, which marks a distinct 
step in the treatment of the subject, but also a useful work of reference." A cademy. 

"This volume is a reproduction, with corrections and additions, of an article 
contributed ^ by the learned author two years ago to the Encyclopedic des Sciences 
Religieuses. It attracted much notice when it first appeared, and is generally 
admitted to present the best summary extant of the vast subject with which it 
deals." Tablet. 

"This is not only on the whole the best but the only manual of the religions of 
India, apart from Buddhism, which we hare in English. The present work 
shows not only great knowledge of the facts and power of clear exposition, but also 
great insight into the inner history and the deeper meaning of the great religion, 
for it is in reality only one, which it proposes to describe." Modern Review. 

" The merit of the work has been emphatically recognised by the most authoritative 
Orientalists, both in this country and on the continent of Europe, But probably 
there are few Indianists (if we may use the word) who would not derive a good deal 
of information from it, and especially from the extensive bibliography provided in 
the notes." Dublin Review. 

" Such a sketch M. Barth has drawn with a master-hand." Critic (New York). 

Post 8vo, pp. viii. 152, cloth, price 6s. 



An Exposition of the System of Kapila, with an Appendix on the 
Nyaya and Vais eshika Systems. 

BY JOHN DAVIES, M.A. (Cantab.), M.R.A.S. 

The system of Kapila contains nearly all that India has produced in the 
department of pure philosophy. 

"The non-Orientalist . . . finds in Mr. Davies a patient and learned guide who 
leads him into the intricacies of the philosophy of India, and supplies him with a clue, 
that he may not be lost in them. In the preface he states that the system of 
Kapila is the earliest attempt 011 record to give an answer, from reason alone, 
to the mysterious questions which arise in every thoughtful mind about the origin of 
the world, the nature and relations of man and his future destiny, and in his learned 
and able notes he exhibits the connection of the Sankhya system with the philo 
sophy of Spinoza, and the connection of the system of Kapila with that of Schopen 
hauer and Von Hartmann. "Foreign Church Chronicle. 

" Mr. Davies s volume on Hindu Philosophy is an undoubted gain to all students 
of the development of thought. The system of Kapila, which is here given in a trans 
lation from the Sankhya Karika. is the only contribution of India to pure philosophy 
. . . Presents many points of deep interest to the student of comparative philo 
sophy, and without Mr. Davies s lucid interpretation it would be difficult to appre 
ciate these points in any adequate manner." Saturday Review. 

" We welcome Mr. Davies s book as a valuable addition to our philosophical 
library." Notts and Queries. 


Second Edition. Post 8vo, pp. x. 130, cloth, price 6s. 


Translated, with copious Annotations, 

Bombay Staff Corps ; Inspector of Army Schools. 

The design of this little work is to provide for missionaries, and for 
others who, like them, have little leisure for original research, an accurate 
summary of the doctrines of the Vedanta. 

" The modest title of Major Jacob s work conveys but an inadequate idea of the 
vast amount of research embodied in his notes to the text of the Vedantasara. So 
copious, indeed, are these, and so much collateral matter do they bring to bear on 
the subject, that the diligent student will rise from their perusal with a fairly 
adequate view of Hindu philosophy generally. His work ... is one of the best of 
its kind that we have seen." Calcutta Review. 

Post 8vo, pp. xii. 154, cloth, price 73. 6d. 




Custodian of the Grey Collection, Cape Town ; Corresponding Member 

of the Geogr. Society, Dresden ; Corresponding Member of the 

Anthropological Society, Vienna, &c. , &c. 

"The first instalment of Dr. Hahn s labours will be of interest, not at the Cape 
only, but in every University of Europe. It is, in fact, a most valuable contribution 
to the comparative study of religion and mythology. Accounts of their religion and 
mythology were scattered about in various books ; these have been carefully col 
lected by Dr. Hahii and printed in his second chapter, enriched and improved by 
what he has been able to collect himself . "Prof. Max Miiller in the Nineteenth 

" It is full of good things." St. James s Gazette. 

hi Four Volumes. Post 8vo, Vol. I., pp. xii. 392, cloth, price 123. 6d., 

Vol. II., pp. vi. 408, cloth, price 123. 6d., Vol. III., pp. viii. 414, 

cloth, price 123. 6d., Vol. IV., pp. viii. 340, cloth, price IDS. 6d. 




Together with a Complete Index to the Text, Preliminary- 
Discourse, and ISTotes. 

By Rev. E. M. WHERRY, M.A., Lodiana. 

" As Mr. Wherry s book is intended for missionaries in India, it is no doubt well 
that they should be prepared to meet, if they can, the ordinary arguments and inter- 
id for this purpose Mr. Wherry s additions will prove useful." Saturday 



Second Edition. Post 8vo, pp. vi. 208, cloth, price 8s. 6d. 


Translated, with Introduction and Notes. 
BY JOHN DA VIES, M.A. (Cantab.) 

" Let us add that his translation of the Bhagavad Gita is, as we judge, the best 
that has as yet appeared in English, and that his Philological Notes are of quite 
peculiar value." Dublin Review. 

Post 8vo, pp. 96, cloth, price 53. 


Translated by E. H. WHINFIELD, M.A., 
Barrister-at-Law, late H.M. Bengal Civil Service. 

Post 8vo, pp. xxxii. 336, cloth, price los. 6d. 


The Persian Text, with an English Verse Translation. 
By E. H. WHINFIELD, late of the Bengal Civil Service. 

" Mr. Whinfield has executed a difficult task with considerable success, and his 
version contains much that will be new to those who only know Mr. Fitzgerald s 
delightful selection. " Academy, 

"The most prominent features in the Quatrains are their profound agnosticism, 
combined with a fatalism based more on philosophic than religious grounds, their 
Epicureanism and the spirit of universal tolerance and charity which animates them." 
Calcutta Review. 

Post 8vo, pp. xxiv. 268, cloth, price 95. 


As exhibited in a series of Articles contributed to the Calcutta Review. 

By ARCHIBALD EDWARD GOUGH, M.A., Lincoln College, Oxford ; 
Principal of the Calcutta Madrasa. 

"For practical purposes this is perhaps the most important of the works that have 
thus far appeared in Triibner s Oriental Series. . . . We cannot doubt that for all 
who may take it up the work must be one of profound interest." Saturday Review. 

In Two Volumes. Vol. I., post 8vo, pp. xxiv. 230, cloth, price 73. 6d. 


By DR. C. P. TIELE. 


Translated from the Dutch with the Assistance of the Author. 


" It places in the hands of the English readers a history of Egyptian Religion 
which is very complete, which is based on the best materials, and which has been 
illustrated by the latest results of research. In this volume there is a great deal of 
information, as well as independent investigation, for the trustworthiness of which 
Dr. Tiele s name is in itself a guarantee ; and the description of the successive 
religions under the Old Kingdom, the Middle Kingdom, and the New Kingdom, is 
given in a manner which is scholarly and minute. " Scotsman. 


Post Svo, pp. xii. 302, cloth, price 8s. 6d. 



Translated from the Persian into English Verse. 

" Mr. Griffith, who has done already good service as translator into verse from the 
Sanskrit, has done farther good work in this translation from the Persian, and he 
has evidently shown not a little skill in his rendering the quaint and very oriental 
style of his author into our more prosaic, less figurative, language. . . . The work, 
besides its intrinsic merits, is of importance as being one of the most popular and 
famous poems of Persia, and that which is read in all the independent native schools 
of India where Persian is taught." Scotsman. 

Post Svo, pp. viii. 266, cloth, price gs. 



" An entirely novel method of dealing with philosophical questions and impart a 
real human interest to the otherwise dry technicalities of the science." Standard. 

" Dr. Abel is an opponent from whom it is pleasant to differ, for he writes with 
enthusiasm arid temper, and his mastery over tfie English language fits him to be a 
champion of unpopular doctrines. Athenaeum. 

Post Svo, pp. ix. 281, cloth, price ics. 6d. 




Translated by E. B. CO WELL, M. A., Professor of Sanskrit in the University 

of Cambridge, and A. E. GOUGH, M.A., Professor of Philosophy 

in the Presidency College, Calcutta. 

This work is an interesting specimen of Hindu critical ability. The 
author successively passes in review the sixteen philosophical systems 
current in the fourteenth century in the South of India ; and he gives what 
appears to him to be their most important tenets. 

"The translation is trustworthy throughout. A protracted sojourn in Indin, 
where there is a living tradition, has familiarised the translators with Indian 
thought." Athenaeum. 

Post Svo, pp. Ixv. 368, cloth, price 143. 


Translated from the Tibetan of the KAH-GYUR. 

Done into English from the German, with an Introduction, 

"Mr. Ralston, whose name is so familiar to all lovers of Russian folk-lore, has 
supplied some interesting Western analogies and parallels, drawn, for the mosc part, 
from Slavonic sources, to the Eastern folk-tales, culled from the Kahgyur, one of the 
divisions of the Tibetan sacred books." Academy. 

"The translation . . . could scarcely have fallen into better hands. An Introduc 
tion . . . gives the leading facts in the lives of those scholars who have given their 
attention to gaining a knowledge of the Tibetan literature and language." Calcutt a 

" Ought to interest all who care for the East, for amusing stories, or for comparative 
folk-lore." Pall Mall Gazette. 


Post 8vo, pp. xvi. 224, cloth, price 93. 



Compiled by DHARMATRATA. 

Translated from the Tibetan of Bkah-hgyur, with Notes, and 
Extracts from the Commentary of Pradjuavarman, 


" Mr. Rockhill s present work is the first from which assistance will be gained 
for a more accurate understanding of the Pali text ; it is, in fact, as yet the only 
term of comparison available to us. The Udanavarga the Thibetan version was 
originally discovered by the late M. Schiefner, who published the Tibetan text, and 
hud intended adding a translation, an intention frustrated by his death, but which 
has been carried out by Mr. Rockhill. ... Mr. Bockhill may be congratulated for 
having well accomplished a difficult task. " Saturday Review. 

In Two Volumes, post 8vo, pp. xxiv. 566, cloth, accompanied by a 
Language Map, price i8s. 


Barrister-at-Law, and late of Her Majesty s Indian Civil Service. 

" Any one at all interested in African languages cannot do better than get Mr. 
Gust s book. It is encyclopaedic in its scope, and the reader gets a start clear away 
in any particular language, and is left free to add to the initial sum of knowledge 
there collected. "Natal Mercury. 

"Mr. Oust has contrived to produce a work of value to linguistic students." 

Third Edition. Post 8vo, pp. xv.-25o, cloth, price js. 6d. 




Doctor of Theology, Professor of the History of Religions in the 
University of Leyden. 

Translated from the Dutch by J. ESTLIN CARPENTER, M.A. 

" Few books of its size contain the result of so much wide thinking, able and labo 
rious study, or enable the reader to gain a better bird s-eye view of the latest results 
of investigations into the religious history of nations. As Professor Tiele modestly 
says, In this little book are outlines pencil sketches, I might say nothing more. 
But there are some men whose sketches from a thumb-nail are of far more worth 
than an enormous canvas covered with the crude painting of others, and it is easy to 
see that these pages, full of information, these sentences, cut and perhaps also dry, 
short arid clear, condense the fruits of long and thorough research. " Scotsman. 


Post 8vo, pp. xii. 312, with Maps and Plan, cloth, price 143. 


Including Burma Proper, Pegu, Taungu, Tenasserim, and Arakan. From 
the Earliest Time to the End of the First War with British India. 

Membre Correspondant de la Societe Academique Indo-Chinoise 

de France. 

"Sir Arthur Phayre s contribution to Triibner s Oriental Series supplies a recog 
nised want, and its appearance has been looked forward to for many years 

General Phayre deserves great credit for the patience and industry which has resulted 
in this History of Burma." Saturday Review. 

Third Edition. Post 8vo, pp. 276, cloth, price ys. 6d. 



Containing a Brief Account of the Three Religions of the Chinese, with 
Observations on the Prospects of Christian Conversion amongst that 

" Dr. Edkins has been most careful in noting the varied and often complex phases 

of opinion, so as to give an account of considerable value of the subject." Scotsman. 

" As a missionary, it has been part of Dr. Edkins duty to study the existing 

religions in China, and his long residence in the country has enabled him to acquire 

an intimate knowledge of them as they at present exist." Saturday Review. 

" Dr. Edkins valuable work, of which this is a second and revised edition, has, 
from the time that it was published, been the standard authority upon the subject 
of which it treats." Nonconformist. 

" Dr. Edkins . . . may now be fairly regarded as among the first authorities on 
Chinese religion and language." British Quarterly Review. 

Post 8vo, pp. X.-274, cloth, price gs. 


Derived from Tibetan Works in the Bkah-hgyur and Bstaii-hgyur. 

Followed by notices on the Early History of Tibet and Khoten. 
Translated by W. W. ROCKHILL, Second Secretary U.S. Legation in China. 

"The volume bears testimony to the diligence and fulness with which the author 
has consulted and tested the ancient documents bearing upon his remarkable sub 
ject." Times. 

" Will be appreciated by those who devote themselves to those Buddhist studies 
which have of late years taken in these Western regions so remarkable a develop 
ment. Its matter possesses a special interest as being derived from ancient Tibetan 
works, some portions of which, here analysed and translated, have not yet attracted 
the attention of scholars. The volume is rich in ancient stories bearing upon the 
world s renovation and the origin of castes, as recorded in these venerable autho 
rities." Daily News. 

Third Edition. Post 8vo, pp. viii.~464, cloth, price i6s. 


With Illustrative Extracts from the Commentaries. 

Translated by J. R. BALLANTYNE, LL.D., late Principal of the Benares 



The work displays a vast expenditure of labour and scholarship, for which 
students of Hindoo philosophy have every reason to be grateful to Dr, Hall and the 
publishers." Calcutta Review. 


In Two Volumes, post 8vo, pp. cviii.-242, and viii.~37o, cloth, price 243. 
Dedicated by permission to H.R.H. the Prince of Wales. 


Translated from the Chinese of Hiuen Tsiang (A.D. 629). 

(Trin. Coll., Camb.) ; R.N. (Retired Chaplain and N.I.) ; Professor of Chinese, 
University College, London ; Rector of Wark, Northumberland, &c. 

An eminent Indian authority writes respecting this work : " Nothing 
more can be done in elucidating the History of India until Mr. Beal s trans 
lation of the Si-yu-ki appears." 

"It is a strange freak of historical preservation that the best account of the con 
dition of India at that ancient period has come down to us in the books of travel 
written by the Chinese pilgrims, of whom Hwen Thsang is the best known." Times, 

Post 8vo, pp. xlviii.~398, cloth, price 123. 


Translated from the Sanskrit, with an Introduction. 
By the late A. C. BURNELL, Ph.D., C.I.E. 

Completed and Edited by E. W. HOPKINS, Ph.D., 
of Columbia College, N.Y. 

" This work is full of interest ; while for the student of sociology and the science 
of religion it is full of importance. Ir is a great boon to get so notable a work in so 
accessible a form, admirably edited, and competently translated." Scotsman. 

"Few men were more competent than Burnell to give us a really good translation 
of this well-known law book, first rendered into English by Sir William Jones. 
Burnell was not only an independent Sanskrit scholar, but an experienced lawyer, 
and he joined to these two important qualifications the rare faculty of being able to 
express bis thoughts in clear and trenchant English. . . . We ought to feel very 
grateful to Dr. Hopkins for having given us all that could be published of the trans 
lation left by Burnell." F. MAX MULLER in the Academy. 

Post 8vo, pp. xii.-234, cloth, price gs. 


Between 1819 and 1842. With a Short Notice of all his Published and Un 
published Works and Essays. From Original and for most part Unpub 
lished Documents. 

By THEODORE DUKA, M.D., F.R.C.S. (Eng.), Surgeon-Major 
H.M. s Bengal Medical Service, Retired, &c. 

" Not too soon have Messrs. Triibner added to their valuable Oriental Series a 
history of the life and works of one of the most gifted and devoted of Oriental 
students, Alexander Csoma de Koros. It is forty-three years since his death, and 
though an account of his career was demanded soon after his decease, it has only 
now appeared in the important memoir of his compatriot, Dr. Duka." Bookseller. 


In Two Volumes, post 8vo, pp. xii.-3i8 and vi.-3i2, cloth, price 2is. 


Reprinted from "Dalrymple s Oriental Repertory," "Asiatic Researches," 
and the "Journal of the Asiatic Society of Bengal." 


I. Some Accounts of Quedah. By Michael Topping. 

II. Report made to the Chief and Council of Balambangan, by Lient. James 
Barton, of his several Surveys. 

HI. Substance of a Letter to the Court of Directors from Mr. John Jesse, dated 
July 20, 1775, at Borneo Proper. 

IV. Formation of the Establishment of Poolo Peenang. 

V. The Gold of Limong. By John Macdonald. 

VI. On Three Natural Productions of Sumatra. By John Macdonald. 

VII. On the Traces of the Hindu Language and Literature extant amongst the 

Malays. By William Marsden. 

VI [I. Some Account of the Elastic Gum Vine of Prince-Wales Island. By James 


IX.-- A Botanical Description of Urceola Elastica, or Caoutchouc Vine of Sumatra 
and Pulo-Pinang. By William Roxburgh, M.D. 

X. An Account of the Inhabitants of the Foggy, or Nassau Islands, lying off 
Sumatra. By John Crisp. 

XI. Remarks on the Species of Pepper which are found on Prince-Wales Island. 
By William Hunter, M.D. 

XII. On the Languages and Literature of the Indo-Chinese Nations. By J. 

Le\den, M.D. 

XIII. Some Account of an Orang-Outang of remarkable height found on the Island 

of Sumatra. By Clarke Abel, M.D. 

XIV. Observations on the Geological Appearances and General Features of Por 
tions of the Malayan Peninsula. By Captain James Low. 

XV. Short Sketch of the Geology of Pulo-Pinang and the Neighbouring Islands. 
By T. Ware. 

XVI. Climate of Singapore. 

XVII. Inscription on the Jetty at Singapore. 

XVIII. Extract of a Letter from Colonel J. Low. 

XIX. Inscription at Singapore. 

XX. An Account of Several Inscriptions found in Province W T ellesley. By Lieut. - 
Col. James Low. 

XXL Note on the Inscriptions from Singapore and Province Wellesley. By J. W. 

XXII. On an Inscription from Keddah. By Lieut.-Col. Low. 

XXIII. A Notice of the Alphabets of the Philippine Islands. 

XXIV. Succinct Review of the Observations of the Tides in the Indian Archipelago. 

XXV. Report on the Tin of the Province of Mergui. By Capt. G. B Tremenheere. 

XXVI. Report on the Manganese of Mergui Province. By Capt. G. B. Tremenheere. 

XXVII. Paragraphs to be added to Capt. G. B. Tremenheere s Report. 

XXVIII.- -Second Report on the Tin of Mergui. By Capt. G. B. Tremenheere. 

XXIX. Analysis of Iron Ores from Tavoy and Mergui, and of Limestone from 
Mergui. By Dr. A. Ure. 

XXX. Report of a Visit to the Pakchan River, and of s< Tin Localities in. the 
Southern Portion of the Tenasserim Provinces. By Capt. G. 13. Tremenheere. 

XXXI. Report on a Route from the Mouth of the Pakchan to Krau, and thence 
across the Isthmus of Krau to the Gulf of Siam. By Capt. Al. Fraser and Capt. J. G. 

XXX 1 1. Report, &c. , from Capt. G. B. Tremenheere on the Price of Mergui Tin Ore. 

XXXIII. Remarks on the Different Species of Orang-utan. By K. Blyth. 

XXXIV. Further Remarks. By E. Blyth. 




XXXV. Catalogue of Mammalia inhabiting the Malayan Peninsula and Islands. 
By Theodore Cantor, M.D. 

XXXVI. On the Local and Relative Geology of Singapore. By J. R. Logan. 

XXX VH. Catalogue of Reptiles inhabiting the Malayan Peninsula and Islands. 
By Theodore Cantor, M.D. 

XXXVI [I. Some Account of the Botanical Collection brought from the Eastward, 
in 1841, by Dr. Cantor. By the late W. Griffith. 

XXXIX. On the Flat-Horned Taurine Cattle of S.E. Asia. By E. Blyth. 

XL. Note, by Major-General G. B. Tremenheere. 

General Index. 

Index of Vernacular Terms. 

Index of Zoological Genera and Sub-Genera occurring in Vol. II. 

"The papers treat of almost every aspect of Indo-China its philology, economy, 
geography, geology and constitute a very material and important contribution to 
our accessible information regarding that country and its people." Contemporary 

Post 8vo, pp. xii.-ya, cloth, price 53. 


Translated from the Sanskrit 


Rector of Eggesford, North Devon. 

" A very interesting addition to Triibner s Oriental Series." Saturday Review. 
" Many of the Maxims in the book have a Biblical ring and beauty of expression." 
St. James Gazette. 

Post 8vo, pp. xii.-i8o, cloth, price 6s. 




Author of "Elements of Pali Grammar," "Translation of the 
Dhammapada, " &c. 

The Sanscrit-Pali word Nlti is equivalent to "conduct" in its abstract, 
and "guide" in its concrete signification. As applied to books, it is a 
general term for a treatise which includes maxims, pithy sayings, and 
didactic stories, intended as a guide to such matters of every-day life as 
form the character of an individual and influence him in his relations to his 
fellow-men. Treatises of this kind have been popular in all ages, and have 
served as a most effective medium of instruction. 

Post 8vo, pp. xxxii. and 330, cloth, price 73. 6d. 



Translated and Abridged by E. H. AVHINFIELD, M.A., 
Late of H.M. Bengal Civil Service. 


Post 8vo, pp. viii. and 346, cloth, price IDS. 6d. 


BY J. JOLLY, Ph.D., 

Professor of Sanskrit in the University of Wurzburg ; late Tagore Professor 

of Law in the University of Calcutta. 

The date assigned by Sir William Jones to this Code the well-known 
Great Law Book of the Hindus is 1250-500 B.C., although the rules and 
precepts contained in it had probably existed as tradition for countless a^es 
before. There has been no reliable edition of the Text for Students for 
many years past, and it is believed, therefore, that Prof. Jolly s work will 
supply a want long felt. 

Post 8vo, pp. 215, cloth, price ys. 6d. 



Author of "Waifs and Strays from the Far East," " Taoist Texts " 
"Idiomatic Phrases in the Peking Colloquial," &c. &c. 

Post 8vo, pp. xvi.-548, with Six Maps, cloth, price 213. 


WRITTEN FROM THE YEAR 1847 TO 1887. Second Scries. 


Barrister-at-Law ; Honorary Secretary of the Royal Asiatic Society 
Late Member of Her Majesty s Indian Civil Service. 

In Two Volumes, post 8vo, pp. x.~3o8 and ^.-314, cloth, price 255 


Edited by R. ROST, Ph.D., &c. &c., 
Librarian to the India Office. 


Reprinted for the Straits Branch of the Royal Asiatic Society from the 
Malayan "Miscellanies," the "Transactions and Journal " of the Bataviau 
Society, and the "Journals" of the Asiatic Society of Bengal and the 
Royal Geographical and Royal Asiatic Societies. 

Post 8vo, pp. xii.-5i2, price i6s. 


By the REV. J. HINTON KNOWLES, F.R.G.S., M.R.A.S, &c. 

(C.M.S.) Missionary to the Kashmiris. 


In Two Volumes, post 8vo, pp. xii.~336 and x.~352, cloth, price 2is. 



Formerly Physician of the Russian Legation at Pekin. 

In Two Volumes, post 8vo, pp. I.~4o8 and 431, cloth, price 363. 




With Notes and Indices by Prof. EDAVARD SACHAU, 

University of Berlin. 

%* The Arabic Original, with an Index of the Sanskrit Words, Edited by 
Professor SACHAU, is in the press. 

Post 8vo, pp. xxxvii.-2i8, cloth, price IDS. 



With a Preface containing an account of the Works of I-TsiNG. 


(Ti!n. Coll., Camb.); Professor of Chinese, University College, London; 
Rector of Wark, Northumberland, kc. 

Author of " Buddhist Records of the Western World," " The Romantic 
Legend of Sakya Budda," &c. 

When the Pilgrim Hiuen Tsiang returned from his travels in India, he 
took up his abode in the Temple of "Great Benevolence ; " this convent had 
been constructed by the Emperor in honour of the Empress, Wen-te-hau. 
After Hiuen Tsiang s death, his disciple, Hwui Li, composed a work which 
gave an account of his illustrious Master s travels ; this work when he com 
pleted he buried, and refused to discover its place of concealment. But 
previous to his death he revealed its whereabouts to Yen-tsung, by whom it 
was finally revised and published. This is " The Life of Hiuen Tsiang." It 
is a valuable sequel to the Si-yu-ki, correcting and illustrating it in many 


Post 8vo. 



Author of " Modern Languages of the East," " Modern Languages of 
Africa," &c. 


250 1/10/89. 











E. B. CO WELL, M.A. 




A. E. GOUGH, M.A. 




[All rights reserved. 1 


I WELL remember the interest excited among the learned 
Hindus of Calcutta by the publication of the Sarva-dar- 
sana-samgraha of Madhava Acharya in the Bibliotheca 
Indica in 1858. It was originally edited by Pandit Isvara- 
chandra Vidyasagara, but a subsequent edition, with no 
important alterations, was published in 1872 by Pandit 
Taranatha Tarkavachaspati. The work had been used by 
Wilson in his " Sketch of the Eeligious Sects of the Hin 
dus " (first published in the Asiatic Eesearches, vol. xvi., 
Calcutta, 1828) ; but it does not appear to have been ever 
much known in India. MS. copies of it are very scarce ; 
and those found in the North of India, as far as I have had 
an opportunity of examining them, seem to be all derived 
from one copy, brought originally from the South, and 
therefore written in the Telugu character. Certain mis 
takes are found in all alike, and probably arose from 
some illegible readings in the old Telugu original. I 
have noticed the same thing in the Nagari copies of 
Madhava s Commentary on the Black Yajur Yeda, which 
are current in the North of India. 

As I was at that time the Oriental Secretary of the Ben- 



gal Asiatic Society, I was naturally attracted to the book ; 
and I subsequently read it with my friend Pandit MaheSa- 
chandra Nyayaratna, the present Principal of the Sanskrit 
College at Calcutta. I always hoped to translate it into 
English; but I was continually prevented by other en 
gagements while I remained in India. Soon after my 
return to England, I tried to carry out my intention ; but 
I found that several chapters, to which I had not paid 
the same attention as to the rest, were too difficult to be 
translated in England, where I could no longer enjoy the 
advantage of reference to my old friends the Pandits of 
the Sanskrit College. In despair I laid my translation 
aside for years, until I happened to learn that my friend, 
Mr. A. E. Gough, at that time a Professor in the Sanskrit 
College at Benares, was thinking of translating the book. 
I at once proposed to him that we should do it together, 
and he kindly consented to my proposal ; and we accord 
ingly each undertook certain chapters of the work. He 
had the advantage of the help of some of the Pandits of 
Benares, especially of Pandit Eama Misra, the assistant 
Professor of Sankhya, who was himself a Eamanuja; 
and I trust that, though we have doubtless left some 
things unexplained or explained wrongly, we may have 
been able to throw light on many of the dark say 
ings with which the original abounds. Our translations 
were originally published at intervals in the Benares 
Pandit between 1874 and 1878; but they have been 
carefully revised for their present republication. 

The work itself is an interesting specimen of Hindu 
critical ability. The author successively passes in review 


the sixteen philosophical systems current in the fourteenth 
century in the South of India, and gives what appeared 
to him to be their most important tenets, and the principal 
arguments by which their followers endeavoured to main 
tain them ; and he often displays some quaint humour as 
he throws himself for the time into the position of their 
advocate, and holds, as it were, a temporary brief in 
behalf of opinions entirely at variance with his own. 1 
We may sometimes differ from him in his judgment of the 
relative importance of their doctrines, but it is always in 
teresting to see the point of view of an acute native critic. 
In the course of his sketches he frequently explains at 
some length obscure details in the different systems ; and I 
can hardly imagine a better guide for the European reader 
who wishes to study any one of these Darsanas in its 
native authorities. In one or two cases (as notably in the 
Bauddha, and perhaps in the Jaina system) he could only 
draw his materials second-hand from the discussions in 
the works of Brahmanical controversialists; but in the 
great majority he quotes directly from the works of their 
founders or leading exponents, and he is continually fol 
lowing in their track even where he does not quote their 
exact words. 2 

The systems are arranged from the Vedanta point of view, 
our author having been elected, in A.D. 1331, the head 

1 The most remarkable instance 2 An index of the names of authors 

of this philosophical equanimity is and works quoted is given in Dr. 

that of Vdchaspati Misra, who wrote Hall s Bibliographical Catalogue, 

standard treatises on each of the six pp. 162-164, and also in Professor 

systems except the Vaiseshika, adopt- Aufrecht s Bodleian Catalogue, p. 

ing, of course, the peculiar point of 247. 
view of each, and excluding for the 
time every alien tenet. 

viii PREFACE. 

of the Srnarta order iu the Math of Sringeri in the 
Mysore territory, founded by Samkara Acharya, the great 
Yedantist teacher of the eighth century, through whose 
efforts the Vedanta became what it is at present the 
acknowledged view of Hindu orthodoxy. The systems 
form a gradually ascending scale, the first, the Charvaka 
and Bauddha, being the lowest as the furthest removed 
from the Vedanta, and the last, the Sankhya and Yoga, 
being the highest as approaching most nearly to it. 

The sixteen systems here discussed attracted to their 
study the noblest minds in India throughout the mediaeval 
period of its history. Hiouen Thsang says of the schools 
in his day : " Les ecoles philosophiques sont constamment 
en lutte, et le bruit de leurs discussions passionnees 
s el&ve comme les flots de la mer. Les heretiques des 
diverses sectes s attachent a des maitres particuliers, et, 
par des voies differentes, marchent tous au meme but." 
We can still catch some faint echo of the din as we read 
the mediaeval literature. Thus, for instance, when King 
Harsha wanders among the Yindhya forests, he finds 
" seated on the rocks and reclining under the trees Arhata 
begging monks, Svetapadas, Mahapasupatas, Pandarabhik- 
shus, Bhagavatas, Yarnins, Kesalufichanas, Lokayatikas, 
Kapilas, Kanadas, Aupanishadas, fsvarakarins, Dharma- 
sastrins, Pauranikas, Saptatantavas, Sabdas, Panchara- 
trikas, &c., all listening to their own accepted tenets and 
zealously defending them." l Many of these sects will 
occupy us in the ensuing pages ; many of them also are 
found in Madhava s poem on the controversial triumphs 

1 Sriharsha-charita, p. 204 (Calcutta ed. ) 


of Samkara Acharya, and in the spurious prose work on 
the same subject, ascribed to Anantanandagiri. Well 
may some old poet have put into the mouth of Yudhish- 
thira the lines which one so often hears from the lips 
of modern pandits 

Veda vibhinnah smritayo vibhimui, 
Nasau munir yasya matam na bhinnam, 
Dharmasya tattvam nihitarn guhayam, 
Mahajano vena gatah sa panthali. 1 

And may we not also say with Clement of Alexandria, 
//,ta<? roivvv ovcrr)? TT}? a\rj6eias, TO jap tyevSos 
eicrpoira^ G^EL, KaOdirep al fid/c^ai TO, TOV HevOews 
prjoraaai, jjue\rj al TT}? 0tXocro0/a? TT}? re /3ap{3dpov TT}? re 
( E\\r)viKfj$ a!p(T6is, eKcia-TT) oirep e\a^ev^ a> 
rrjv d\r)0eiav, ^QJTO? S , ol/uai, dvaro\fj Trdvra (fr 

E. B. C. 

1 Found in the Mahabh. iii. 17402, with some variations. I give them 
as I have heard them from Pandit Ra mana ra yaria Vidyaratna. 



I. The Charvaka System (E. B. C.) 2 

II. The Bauddha System (A. E. G.) . . . . . 12 

III. The Arhata or Jaina System (E. B. C.) . . . . 36 

IV. The Ramdnuja System (A. E. G.) . . . . . 64 
V. The Purna-prajna System (A. E. G.) . . . . 87 

VI. The Nakulfsa-Pasupata System (A. E. G.) . . . . 103 

VII. The Saiva System (E. B. C.) 112 

VIII. The Pratyabhijna or Recognitive System (A. E. G.) . 128 

IX. The RasesVara or Mercurial System (A. E. G.) . .137 

X. The Vaifeshika or Aulukya System (E. B. C.) . . 145 

XL The Akshapada or Nyaya System (E. B. C.) . . 161 

XII. The Jaiminiya System (E. B. C.) . . . . .178 

XIII. The Paniniya System (E. B. C.) 203 

XIV. The Sankhya System (E. B. C.) 221 

XV. The Patanjala or Yoga System (E. B. C.) . . . 231 

XVI. The Vedanta or System of Samkara Acharya . .273 
APPENDIX On the Upddhi (E. B. C.) . . . 275 



1. I worship Siva, the abode of eternal knowledge, the 
storehouse of supreme felicity; by whom the earth and 
the rest were produced, in him only has this all a maker. 

2. Daily I follow my Guru Sarvajna- Vishnu, who knows 
all the Agamas, the son of Sarngapani, who has gone to 
the further shore of the seas of all the systems, and has 
contented the hearts of all mankind by the proper mean 
ing of the term Soul. 

3. The synopsis of all the systems is made by the vener 
able Madhava, mighty in power, the Kaustubha-jewel of 
the milk-ocean of the fortunate Sayana. 

4. Having thoroughly searched the Sastras of former 
teachers, very hard to be crossed, the fortunate Sayana- 
Madhava 1 the lord has expounded them for the delight of 
the good. Let the virtuous listen with a mind from which 
all envy has been far banished ; who finds not delight in 
a garland strung of various flowers ? 

1 Dr. A. C. Burnell, in his preface description of his body, himself being 

to his edition of the Vamsa-Brah- the eternal soul. His use of the 

mana, has solved the riddle of the term Sayana-Miidhavah here (not 

relation of Mddhava and Sayana. the dual) seems to prove that the two 

Siiyana is a pure Dravidian name names represent the same person, 

given to a child who is born after all The body seems meant by the Sayana 

the elder children have died. Ma- of the third sloka. M<iyana was the 

dhava elsewhere calls Sdyana his father of Madhava, and the true 

" younger brother," as an allegorical reading may be sriman-mdyana^ 




[\\ 7 E have said in our preliminary invocation " salutation 
to Siva, the abode of eternal knowledge, the storehouse of 
supreme felicity,"] but how can we attribute to the Divine 
Being the giving of supreme felicity, when such a notion 
has been utterly abolished by Charvaka, the crest-gem of 
the atheistical school, the follower of the doctrine of 
Lrihaspati ? The efforts of Charvaka are indeed hard to 
be eradicated, for the majority of living beings hold by the 
current refrain 

While life is yours, live joyously ; 
None can escape Death s searching eve : 
When once this frame of ours they burn, 
How shall it e er again return ? 

The mass of men, in accordance with the Sastras of 
policy and enjoyment, considering wealth and desire the 
only ends of man, and denying the existence of any object 
belonging to a future world, are found to follow only the 
doctrine of Charvaka. Hence another name for that 
school is Lokayata, a name well accordant with the 
thing signified. 1 

In this school the four elements, earth, &c., are the 

1 " Saiikara, Bhdskara, and other etymologically analysed as " preva- 

commentators name the Lokaya- lent in the world " (loJca and dyata). 

tikas, and these appear to be a Laukayatika occurs in Panini s uk- 

branch of the Sect of Chdrvaka " thaganu. 
(Colebrooke). Lokayata may be 


original principles; from these alone, when transformed 
into the body, intelligence is produced, just as the in 
ebriating power is developed from the mixing of certain 
ingredients ; l and when these are destroyed, intelligence at 
once perishes also. They quote the Sruti for this [Brihad 
Arany. Up. ii. 4, 12], "Springing forth from these ele 
ments, itself solid knowledge, it is destroyed when they 
are destroyed, after death no intelligence remains." * 
Therefore the soul is only the body distinguished by the 
attribute of intelligence, since there is no evidence for any 
soul distinct from the body, as such cannot be proved, 
since this school holds that perception is the only source 
of knowledge and does not allow inference, &c. 

The only end of man is enjoyment produced by sensual 
pleasures. Nor may you say that such cannot be called 
the end of man as they are always mixed with some kind 
of pain, because it is our wisdom to enjoy the pure plea 
sure as far as we can, and to avoid the pain which inevi 
tably accompanies it; just as the man who desires fish 
takes the fish with their scales and bones, and having 
taken as many as he wants, desists ; or just as the man 
who desires rice, takes the rice, straw and all, and having 
taken as much as he wants, desists. It is not therefore 
for us, through a fear of pain, to reject the pleasure which 
our nature instinctively recognises as congenial. Men do 
not refrain from sowing rice, because forsooth there are 
wild animals to devour it ; nor do they refuse to set the 
cooking-pots on the fire, because forsooth there are beggars 
to pester us for a share of the contents. If any one were 

1 Kinwa is explained as "drug or chewed together have an exhilara- 
seed used to produce fermentation ting property not found in those 
in the manufacture of spirits from substances severally." 
sugar, bassia, &c." Colebrooke a Of course Sankara, in his corn- 
quotes from Saiikara : "The faculty mentary, gives a very different in- 
of thought results from a modifica- terpretation, applying it to the cessa 
tion of the aggregate elements in tion of individual existence when the 
like manner as sugar with a ferment knowledge of the Supreme is once 
and other ingredients becomes an attained. Cf. Sabara s Coinm. Jai- 
inebriating liquor ; and as betel, mini Sut., i. i. 5. 
areca, lime, and extract of catechu 


so timid as to forsake a visible pleasure, he would indeed 
be foolish like a beast, as has been said by the poet 

The pleasure which arises to men from contact with sensible objects, 
Is to be relinquished as accompanied by pain, such is the reasoning 

of fools ; 

The berries of paddy, rich with the finest white grains, 
What man, seeking his true interest, would fling away because 

covered with husk and dust 1 l 

If you object that, if there be no such thing as happi 
ness in a future world, then how should men of experienced 
wisdom engage in the agnihotra and other sacrifices, which 
can only be performed with great expenditure of money 
and bodily fatigue, your objection cannot be accepted 
as any proof to the contrary, since the agnihotra, &c., are 
only useful as means of livelihood, for the Veda is tainted 
by the three faults of untruth, self-contradiction, and tau 
tology; 2 then again the impostors who call themselves 
Vaidic pundits are mutually destructive, as the authority 
of the jmina-kanda is overthrown by those who maintain 
that of the karma-kanda, while those who maintain the 
authority of the jnana-kanda reject that of the karma- 
kanda ; and lastly, the three Vedas themselves are only 
the incoherent rhapsodies of knaves, and to this effect runs 
the popular saying 

The Agnihotra, the three Vedas, the ascetic s three staves, and smear 
ing oneself with ashes, 

Brihaspati says, these are but means of livelihood for those who have 
no manliness nor sense. 

Hence it follows that there is no other hell than mun 
dane pain produced by purely mundane causes, as thorns, 
&c. ; the only Supreme is the earthly monarch whose 
existence is proved by all the world s eyesight ; and the 
only Liberation is the dissolution of the body. By hold 
ing the doctrine that the soul is identical with the body, 

1 I take kana as here equal to the Bengali Tcunr. Cf. Atharva-V., xi. 
3, 5. Asvdh kand gdvas tcmduld masakds tushdh. 
2 See Nyaya Sutras, ii. 57. 


such phrases as " I am thin," " I am black," &c., are at 
once intelligible, as the attributes of thinness, &c., and self- 
consciousness will reside in the same subject [the body] ; 
like and the use of the phrase " my body " is metaphorical 
" the head of Kahu " [Kahu being really all head]. 
All this has been thus summed up 

In this school there are four elements, earth, water, fire, and air ; 
And from these four elements alone is intelligence produced, 
Just like the intoxicating power from kinwa, &c., mixed together ; 
Since in "I am fat," "I am lean," these attributes 1 abide in the 

same subject, 
And since fatness, &c., reside only in the body, 2 it alone is the soul 

and no other, 
And such phrases as "my body " are only significant metaphorically. 

" Be it so," says the opponent ; " your wish would be 
gained if inference, &c., had no force of proof ; but then 
they have this force ; else, if they had not, then how, on 
perceiving smoke, should the thoughts of the intelligent 
immediately proceed to fire ; or why, on hearing another 
say, There are fruits on the bank of the river, do those 
who desire fruit proceed at once to the shore ? " 

All this, however, is only the inflation of the world of 

Those who maintain the authority of inference accept 
the sign or middle term as the causer of knowledge, which 
middle term must be found in the minor and be itself 
invariably connected with the major. 3 Now this invariable 
connection must be a relation destitute of any condition 
accepted or disputed; 4 and this connection does not possess 
its power of causing inference by virtue of its existence, as 
the eye, &c., are the cause of perception, but by virtue of 
its being known. What then is the means of this con 
nection s bein^ known ? 


1 7.e., personality and fatness, &c. 4 For the sandigdha and ntichifa 

2 I read dehe for dehah. upddhi see Siddhanta MuktaVali, p. 

3 Literally, " must be an attribute 125. The former is accepted only 
of the subject and have invariable by one party. 

concomitance (rydpti)." 


We will first show that it is not perception. Now per 
ception is held to be of two kinds, external and internal 
[i.e., as produced by the external senses, or by the inner 
sense, mind]. The former is not the required means ; for 
although it is possible that the actual contact of the 
senses and the object will produce the knowledge of the 
particular, object thus brought in contact, yet as there can 
never be such contact in the case of the past or the future, 
the universal proposition 1 which was to embrace the in 
variable connection of the middle and major terms in 
every case becomes impossible to be known. Nor may 
you maintain that this knowledge of the universal pro 
position has the general class as its object, because if so, 
there might arise a doubt as to the existence of the inva 
riable connection in this particular case 2 [as, for instance, 
in this particular smoke as implying fire]. 

Nor is internal perception the means, since you cannot 
establish that the mind has any power to act indepen 
dently towards an external object, since all allow that it 
is dependent on the external senses, as has been said by 
one of the logicians, " The eye, &c., have their objects as 
described; but mind externally is dependent on the 

Nor can inference be the means of the knowledge of the 
universal proposition, since in the case of this inference 
we should also require another inference to establish it, 
and so on, and hence would arise the fallacy of an ad 
infinitum retrogression. 

Nor can testimony be the means thereof, since we may 
either allege in reply, in accordance with the Vaiseshika 
doctrine of Kanada, that this is included in the topic of 
inference ; or else we may hold that this fresh proof of 
testimony is unable to leap over the old barrier that 

1 Literally, the knowledge of the thus idiots are men, though man 
invariable concomitance (as of smoke is a rational animal ; and again, this 
by fire). particular smoke might be a sign of 

2 The attributes of the class are a fire in some other place, 
not always found in every member, 


stopped the progress of inference, since it depends itself 
on the recognition of a sign in the form of the language 
used in the child s presence by the old man ; l and, more 
over, there is no more reason for our believing on another s 
word that smoke and fire are invariably connected, than 
for our receiving the ipse dixit of Manu, &c. [which, of 
course, we Charvakas reject]. 

And again, if testimony were to be accepted as the only 
means of the knowledge of the universal proposition, then 
in the case of a man to whom the fact of the invariable 
connection between the middle and major terms had not 
been pointed out by another person, there could be no 
inference of one thing [as fire] on seeing another thing [as 
smoke] ; hence, on your own showing, the whole topic of 
inference for oneself 2 would have to end in mere idle 

Then again comparison? &c., must be utterly rejected as 
the means of the knowledge of the universal proposition, 
since it is impossible that they can produce the knowledge 
of the unconditioned connection [i.e., the universal pro 
position], because their end is to produce the knowledge of 
quite another connection, viz., the relation of a name to 
something so named. 

Again, this same absence of a condition, 4 which has been 
given as the definition of an invariable connection [i.e., a 
universal proposition], can itself never be known ; since it 
is impossible to establish that all conditions must be objects 
of perception ; and therefore, although the absence of per- 

1 See Sahitya Darpana (Ballan- named." Ballantyue s Tarka San- 
tyne s trans, p. 16), and Siddhanta- graha. 

M., p. 80. 4 The upa dhi is the condition which 

2 The properly logical, as distin- must be supplied to restrict a too 
guished from the rhetorical, argu- general middle term, as in the in- 
ment. ference "the mountain has smoke 

3 " Upamdna or the knowledge of because it has fire," if we add wet 
a similarity is the instrument in the fuel as the condition of the fire, the 
production of an inference from middle term will be no longer too 
similarity. This particular inference general. In the case of a true vyapti, 
consists in the knowledge of the there is, of course, no upadhi. 
relation of a name to something so 


ceptible things may be itself perceptible, the absence of 
non-perceptible things must be itself non-perceptible ; and 
thus, since we must here too have recourse to inference, 
&c., we cannot leap over the obstacle which has already 
been planted to bar them. Again, we must accept as the 
definition of the condition, " it is that which is reciprocal 
or equipollent in extension 1 with the major term though 
not constantly accompanying the middle." These three 
distinguishing clauses, " not constantly accompanying the 
middle term," " constantly accompanying the major term," 
and "being constantly accompanied by it " [i.e., reciprocal], 
are needed in the full definition to stop respectively three 
such fallacious conditions, in the argument to prove the 
non-eternity of sound, as " being produced," " the nature 
of a jar," and "the not causing audition;" 2 wherefore the 
definition holds, and again it is established by the sloka 
of the great Doctor beginning samdsama. 3 

wherever the class of jar is found 
there is also found non-eternity. 
Lastly, if we defined the upadhi as 
"not constantly accompanying the 
middle term, and constantly accom 
panying the major," we might have 
as a Mimdmsaka upddhi "the not 
causing audition," i.e., the not being 
apprehended by the organs of hear 
ing ; but this is excluded, as non-eter 
nity is not always found where this 
is, ether being inaudible and yet 

3 This refers to an obscure sloka 
of Udayana charya, " where a recip 
rocal and a non-reciprocal universal 
connection (i.e., universal proposi 
tions which severally do and do not 
distribute their predicates) relate to 
the same argument (as e.g., to prove 
the existence of smoke), there that 
non-reciprocating term of the second 
will be a fallacious middle, which is 
not invariably accompanied by the 
other reciprocal of the first." Thus 
" the mountain has smoke because it 
has fire " (here fire and smoke are 
non-reciprocating, as fire is not found 
invariably accompanied by smoke 

(Pr. Anal., ii. 25). 
We have here our A with distributed 

2 If we omitted the first clause, 
and only made the upddhi "that which 
constantly accompanies the major 
term and is constantly accompanied 
by it," then in the Naiya"yika argu 
ment " sound is non-eternal, because 
it has the nature of sound," "being 
produced " would serve as a Mimdm- 
saka upa"dhi, to establish the vya- 
Ihichdra fallacy, as it is reciprocal 
with "non-eternal ;" but the omitted 
clause excludes it, as an uptldhi 
must be consistent with either party s 
opinions, and, of course, the Naiya"- 
yika maintains that "being pro 
duced " ahvays accompanies the class 
of sound. Similarly, if we defined 
the upsldhi as "not constantly accom 
panying the middle term and con 
stantly accompanied by the major," 
we might have as an upddhi " the 
nature of a jar," as this is never 
found with the middle term (the 
class or nature of sound only resid 
ing in sound, and that of a jar only 
in a jar), while, at the same time, 


But since the knowledge of the condition must here 
precede the knowledge of the condition s absence, it is 
only when there is the knowledge of the condition, that 
the knowledge of the universality of the proposition is 
possible, i.e., a knowledge in the form of such a connection 
between the middle term and major term as is distinguished 
by the absence of any such condition ; and on the other 
hand, the knowledge of the condition depends upon the 
knowledge of the invariable connection. Thus we fasten 
on our opponents as with adamantine glue the thunder 
bolt-like fallacy of. reasoning in a circle. Hence by the 
impossibility of knowing the universality of a proposition 
it becomes impossible to establish inference, &C. 1 

The step which the mind takes from the knowledge of 
smoke, &c., to the knowledge of fire, &c., can be accounted 
for by its being based on a former perception or by its 
being an error ; and that in some cases this step is justified 
by the result, is accidental just like the coincidence of 
effects observed in the employment of gems, charms, 
drugs, &c. 

From this it follows that fate, &c., 2 do not exist, since 
these can only be proved by inference. But an opponent 
will say, if you thus do not allow adrishta, the various 
phenomena of the world become destitute of any cause. 

though smoke is by fire), or "because which is the reciprocal of fire. I 

it has fire from wet fuel " (smoke and wish to add here, once for all, that 

fire from wet fuel being reciprocal I own my explanation of this, as 

and always accompanying each well as many another, difficulty 

other) ; the non-reciprocating term in the Sarva-darsana-sangraha to 

of the former (fire) will give a falla- my old friend and teacher, Pandit 

cious inference, because it is also, of Mahesa Chandra Nyayaratna, of the 

course, not invariably accompanied Calcutta Sanskrit College, 
by the special kind of fire, that pro- 1 Cf. Sextus Empiricus, P. Hyp. 

duced from wet fuel. But this will ii. In the chapter on the Buddhist 

not be the case where the non-re- system infra, we have an attempt 

ciprocating term is thus invariably to establish the authority of the 

accompanied by the other reciprocal, universal proposition from the rela- 

as " the mountain has fire because it tion of cause and effect or genus and 

has smoke ; " here, though fire and species. 

smoke do not reciprocate, yet smoke 2 Adrishta, i.e., the merit and de- 
will be a true middle, because it is merit in our actions which produce 
invariably accompanied by heat, their effects in future births. 


But we cannot accept this objection as valid, since 
these phenomena can all be produced spontaneously 
from the inherent nature of things. Thus it has been 

The fire is hot, the water cold, refreshing cool the breeze of morn ; 
By whom came this variety ? from their own nature was it born. 

And all this has been also said by Brihaspati 

There is no heaven, no final liberation, nor any soul in another 

Nor do the actions of the four castes, orders, &c., produce any real 

The Agnihotra, the three Yedas, the ascetic s three staves, and smear 
ing one s self with ashes, 

Were made by Nature as the livelihood of those destitute of know 
ledge and manliness. 

If a beast slain in the Jyotishtoma rite will itself go to heaven, 

Why then does not the sacrificer forthwith offer his own father ? l 

If the Sraddha produces gratification to beings who are dead, 

Then here, too, in the case of travellers when they start, it is needless 
to give provisions for the journey. 

If beings in heaven are gratified by our offering the Sraddha here, 

Then why not give the food down below to those who are standing 
on the housetop ? 

While life remains let a man live happily, let him feed on ghee even 
though he runs in debt ; 

When once the body becomes ashes, how can it ever return again ? 

If he who departs from the body goes to another world, 

How is it that he comes not back again, restless for love of his 
kindred ? 

Hence it is only as a means of livelihood that Brahmans have estab 
lished here 

All these ceremonies for the dead, there is no other fruit any 

The three authors of the Vedas were buffoons, knaves, and demons. 

All the well-known formulae of the pandits, jarphari, turphari, &c. 2 

And all the obscene rites for the queen commanded in the As wa 
rned ha, 

1 This is an old Buddhist retort. Aswamedha rites, see Wilson s Rig- 
See Burnouf, Introd., p. 209. Veda, Preface, vol. ii. p. xiii. 

2 Rig -Veda, x. 106. For the 


These were invented by buffoons, and so all the various kinds of pre 
sents to the priests, 1 

While the eating of flesh was similarly commanded by night-prowling 

Hence in kindness to the mass of living beings must we 
fly for refuge to the doctrine of Charvaka. Such is the 
pleasant consummation. E. B. C. 

1 Or this may mean "and all the, various other things to be handled in 
the rites." 



AT this point the Buddhists remark: As for what you 
(Charvakas) laid down as to the difficulty of ascertaining 
invariable concomitance, your position is unacceptable, 
inasmuch as invariable concomitance is easily cognisable 
by means of identity and causality. It has accordingly 
been said 

" From the relation of cause and effect, or from identity 
as a determinant, results a law of invariable con 
comitance not through the mere observation of 
the desired result in similar cases, nor through the 
non-observation of it in dissimilar cases." 1 
On the hypothesis (of the Naiyayikas) that it is con 
comitance and non-concomitance (e.g., A is where B is, 
A is not where B is not) that determine an invariable 
connection, the unconditional attendance of the major 
or the middle term would be unascertainable, it being 
impossible to exclude all doubt with regard to in 
stances past and future, and present but unperceived. 
If one (a Naiyayika) rejoin that uncertainty in regard to 
such instances is equally inevitable on our system, we 
reply : Say not so, for such a supposition as that an effect 
may be produced without any cause would destroy itself 
by putting a stop to activity of any kind ; for such doubts 

1 This sloka is quoted in the the second line is there read more 
" Benares Pandit," vol. i. p. 89, with correctly, darsandn tui na darsandt. 
a commentary, and the latter part of 


alone are to be entertained, the entertainment of which 
does not implicate us in practical absurdity and the like, 
as it has been said, " Doubt terminates where there is a 
practical absurdity." l 

1. By ascertainment of an effectuation, then, of that (viz., 
of the designate of the middle) is ascertained the invariable 
concomitance (of the major) ; and the ascertainment of 
such effectuation may arise from the well-known series of 
five causes, in the perceptive cognition or non-cognition of 
cause and effect. That fire and smoke, for instance, stand 
in the relation of cause and effect is ascertained by five 
indications, viz., (i.) That an effect is not cognised prior 
to its effectuation, that (2.) the cause being perceived (3.) 
the effect is perceived, and that after the effect is cognised 
(4.) there is its non-cognition, (5.) when the (material) 
cause is no longer cognised. 

2. In like manner an invariable concomitance is ascer 
tained by the ascertainment of identity (e.g., a sisu-tree is 
a tree, or wherever we observe the attributes of a sisu we 
observe also the attribute arboreity), an absurdity attach 
ing to the contrary opinion, inasmuch as if a sisu-tree 
should lose its arboreity it would lose its own self. But, 
on the other hand, where there exists no absurdity, and 
where a (mere) concomitance is again and again observed, 
who can exclude all doubt of failure in the concomitance ? 
An ascertainment of the identity of sisu and tree is com 
petent in virtue of the reference to the same object (i.e., 
predication), This tree is a sisu. For reference to the 
same object (predication) is not competent where there is 
no difference whatever (e.g., to say, " A jar is a jar," is no 
combination of diverse attributes in a common subject), 
because the two terms cannot, as being synonymous, be 
simultaneously employed ; nor can reference to the same 
object take place where there is a reciprocal exclusion (of 
the two terms), inasmuch as we never find, for instance, 
horse and cow predicated the one of the other. 

1 Kusumanjali, iii. 7. 


It has thus been evinced that an effect or a self-same 
supposes a cause or a self-same (as invariable concomi 

If a man does not allow that inference is a form of 
evidence, pramdna, one may reply : You merely assert thus 
much, that inference is not a form of evidence : do you 
allege no proof of this, or do you allege any ? The former 
alternative is not allowable according to the maxim that 
bare assertion is no proof of the matter asserted. Nor is 
the latter alternative any better, for if while you assert 
that inference is no form of evidence, you produce some 
truncated argument (to prove, i.e., infer, that it is none), 
you will be involved in an absurdity, just as if you asserted 
your own mother to be barren. Besides, when you affirm 
that the establishment of a form of evidence and of the 
corresponding fallacious evidence results from their homo 
geneity, you yourself admit induction by identity. Again, 
when you affirm that the dissentiency of others is known 
by the symbolism of words, you yourself allow induction 
by causality. When you deny the existence of any object 
on the ground of its not being perceived, you yourself 
admit an inference of which non-perception is the middle 
term. Conformably it has been said by Tathagata 

" The admission of a form of evidence in general results 
from its being present to the understanding of 

" The existence of a form of evidence also follows from 
its negation by a certain person." 

All this has been fully handled by great authorities; 
and we desist for fear of an undue enlargement of our 

These same Bauddhas discuss the highest end of man 
from four standpoints. Celebrated under the designations 
of Madhyamika, Yogachara, Sautrantika, and Vaibhashika, 
these Buddhists adopt respectively the doctrines of a 
universal void (nihilism), an external void (subjective 
idealism), the inferribility of external objects (representa- 


tionism), and the perceptibility of external objects (pre- 
sentationism). 1 Though the venerated Buddha be the only 
one teacher (his disciples) are fourfold in consequence of 
this diversity of views; just as when one has said, "The 
sun has set," the adulterer, the thief, the divinity student, 
and others understand that it is time to set about their 
assignations, their theft, their religious duties, and so forth, 
according to their several inclinations. 

It is to be borne in mind that four points of view have 
been laid out, viz., (i.) All is momentary, momentary; (2.) 
all is pain, pain; (3.) all is like itself alone; (4.) all is 
void, void. 

Of these points of view, the momentariness of fleeting 
things, blue and so forth (i.e., whatever be their quality), 
is to be inferred from their existence ; thus, whatever is 
is momentary (or fluxional) like a bank of clouds, and all 
these things are. 2 Nor may any one object that the 
middle term (existence) is unestablished ; for an existence 
consisting of practical efficiency is established by percep 
tion to belong to the blue and other momentary things ; 
and the exclusion of existence from that which is not 
momentary is established, provided that we exclude from 

1 The Bauddhas are thus divided is that ? That conclusion is that 

into you never, even for the shortest time 

(i.) Mddhyamikas or Nihilists. that can be named or conceived, see 

(2.) Yogdcharas or Subjective any abiding colour, any colour which 

Idealists. truly is. Within the millionth part 

(3.) Sautrdntikas or Representa- of a second the whole glory of the 

tionists. painted heavens has undergone an 

(4.) Vaibhdshikas or Presenta- incalculable series of mutations. One 

tionists. shade is supplanted by another with 

a Cf. Ferrier s Lectures and Re- a rapidity which sets all measure- 
mains, vol. i. p. 119. ment at defiance, but because the 

" Suppose yourself gazing on a process is one to which no measure- 
gorgeous sunset. The whole western ment applies, . . . reason refuses 
heavens are glowing with roseate to lay an arrestment on any period 
hues, but you are aware that with- of the passing scene, or to declare 
in half an hour all these glorious that it is, because in the very act of 
tints will have faded away into a being it is not ; it has given place to 
dull ashen grey. You see them even something else. It is a series of 
now melting away before your eyes, fleeting colours, no one of which is, 
although your eyes cannot place be- because each of them continually 
fore you the conclusion which your vanishes in another." 
reason draws. And what conclusion 


it the non-momentary succession and simultaneity, accord 
ing to the rule that exclusion of the continent is exclusion 
of the contained. Now this practical efficiency (here 
identified with existence) is contained under succession 
and simultaneity, and no medium is possible between 
succession and non-succession (or simultaneity) ; there 
being a manifest absurdity in thinking otherwise, accord 
ing to the rule 

" In a reciprocal contradiction there exists no ulterior 
alternative ; 

"Nor is their unity in contradictories, there being a 
repugnance in the very statement." l 

And this succession and simultaneity being excluded 
from the permanent, and also excluding from the per 
manent all practical efficiency, determine existence of the 
alternative of momentariness. Q.E.D. 

Perhaps some one may ask: Why may not practical 
efficiency reside in the non-fluxional (or permanent) ? If 
so, this is wrong, as obnoxious to the following dilemma. 
Has your "permanent" a power of past and future practical 
efficiency during its exertion of present practical efficiency 
or no ? On the former alternative (if it has such power), 
it cannot evacuate such past and future efficiency, because 
we cannot deny that it has power, and because we infer 
the consequence, that which can at any time do anything 
does not fail to do that at that time, as, for instance, a com 
plement of causes, and this entity is thus powerful. On the 
latter alternative (if the permanent has no such power of 
past and future agency), it will never do anything, because 
practical efficiency results from power only ; what at any 
time does not do anything, that at that time is unable to 
do it, as, for instance, a piece of stone does not produce a 
germ ; and this entity while exerting its present practical 
efficiency, does not exert its past and future practical 
efficiency. Such is the contradiction. 

You will perhaps rejoin : By assuming successive sub- 

1 Principium exclusi medii inter duo contradictoria. 


sidiaries, there is competent to the permanent entity a 
successive exertion of past and future practical efficiency. 
If so, we would ask you to explain : Do the subsidiaries 
assist the entity or not? If they do not, they are not 
required ; for if they do nothing, they can have nothing 
to do with the successive exertion. If they do assist the 
thing, is this assistance (or supplementation) other than 
the thing or not ? If it is other than the thing, then this 
adscititious (assistance) is the cause, and the non-momen 
tary entity is not the cause : for the effect will then follow, 
by concomitance and non-concomitance, the adventitious 
supplementation. Thus it has been said : 

" What have rain and shine to do with the soul ? Their 
effect is on the skin of man ; 

" If the soul were like the skin, it would be non-perma 
nent ; and if the skin were like the soul, there could 
be no effect produced upon it." 

Perhaps you will say: The entity produces its effect, 
together with its subsidiaries. Well, then (we reply), let 
the entity not give up its subsidiaries, but rather tie them 
lest they fly with a rope round their neck, and so produce 
the effect which it has to produce, and without forfeiting 
its own proper nature. Besides (we continue), does the 
additament (or supplementation) constituted by the sub 
sidiaries give rise to another additament or not ? In 
either case the afore-mentioned objections will coine down 
upon you like a shower of stones. On the alternative 
that the additament takes on another additament, you will 
be embarrassed by a many-sided regress in infinitum. If 
when the additament is to be generated another auxiliary 
(or additament) be required, there will ensue an endless 
series of such additaments : this must be confessed to be 
one infinite regress. For example, let a seed be granted 
to be productive when an additament is given, consisting 
of a complement of objects such as water, wind, and the 
like, as subsidiaries ; otherwise an additament would be 
manifested without subsidiaries. Now the seed in taking 


on the additament takes it on with the need of (ulterior) 
subsidiaries ; otherwise, as there would always be sub 
sidiaries, it would follow that a germ would always be 
arising from the seed. We shall now have to add to the 
seed another supplementation by subsidiaries themselves 
requiring an additament. If when this additament is 
given, the seed be productive only on condition of sub 
sidiaries as before, there will be established an infinite 
regression of additaments to (or supplementations of) the 
seed, to be afforded by the subsidiaries. 

Again, we ask, does the supplementation required for 
the production of the effect produce its effect independently 
of the seed and the like, or does it require the seed and 
the like ? On the first alternative (if the supplementation 
works independently), it would ensue that the seed is in 
no way a cause. On the second (if the supplementation 
require the seed), the seed, or whatever it may be that is 
thus required, must take on a supplementation or addita 
ment, and thus there will be over and over again an end 
less series of additaments added to the additament con 
stituted by the seed ; and thus a second infinite regression 
is firmly set up. 

In like manner the subsidiary which is required will 
add another subsidiary to the seed, or whatever it may be 
that is the subject of the additions, and thus there will be 
an endless succession of additaments added to the addita 
ments to the seed which is supplemented by the sub 
sidiaries; and so a third infinite regression will add to 
your embarrassment. 

Now (or the other grand alternative), let it be granted 
that a supplementation identical with the entity (the seed, 
or whatever it may be) is taken on. If so, the former 
entity, that minus the supplementation, is no more, and a 
new entity identical with the supplementation, and desig 
nated (in the technology of Buddhism) kurvad rupa (or 
elfect-producing object), comes into being : and thus the 


tree of my desires (my doctrine of a universal flux) lias 
borne its fruit. 

Practical efficiency, therefore, in the non-momentary is 
inadmissible. Nor is practical efficiency possible apart 
from succession in time ; for such a possibility is redargued 
by the following dilemma. Is this (permanent) entity 
(which you contend for) able to produce all its effects 
simultaneously, or does it continue to exist after produc 
tion of effects ? On the former alternative, it will result 
that the entity will produce its effects just as much at one 
time as at another; on the second alternative, the expecta 
tion of its permanency is as reasonable as expecting seed 
eaten by a mouse to germinate. 

That to which contrary determinations are attributed is 
diverse, as heat and cold ; but this thing is determined by 
contrary attributions. Such is the argumentation applied 
to the cloud (to prove that it has not a permanent but a 
fluxional existence). Nor is the middle term disallowable, 
for possession and privation of power and impotence are 
allowed in regard to the permanent (which you assert) at 
different times. The concomitance and non-concomitance 
already described (viz., That which can at any time do 
anything does not fail to do that at that time, and What 
at any time does not do anything, that at that time is 
unable to do it) are affirmed (by us) to prove the existence 
of such power. The negative rule is : What at any time 
is unable to produce anything, that at that time does not 
produce it, as a piece of stone, for example, does not pro- 
iduce a germ; and this entity (the seed, or whatever it 
may be), while exerting a present practical efficiency, is 
incapable of past and future practical efficiencies. The 
contradiction violating this rule is : What at any time 
does anything, that at that time is able to do that 
thing, as a complement of causes is able to produce its 
effect; and this (permanent) entity exerts at time past 
and time future the practical efficiencies proper to those 


(To recapitulate.) Existence is restricted to the momen 
tary ; there being observed in regard to existence a nega 
tive rule, that in regard to permanent succession and 
simultaneity being excluded, existence which contains 
succession and simultaneity is not cognisable ; and there 
being observed in regard to existence a positive rule, in 
virtue of a concomitance observed (viz., that the existent 
is accompanied or "pervaded" by the momentary), and 
in virtue of a non-concomitance observed (viz., that the 
non-momentary is accompanied or "pervaded" by the 
non-existent). Therefore it has been said by Jnana-sri 
" What is is momentary, as a cloud, and as these existent 

things ; 

" The power of existence is relative to practical efficiency, 
and belongs to the ideal ; but this power exists not 
as eternal in things eternal (ether, &c.) ; 
" jSTor is there only one form, otherwise one thing could 

do the work of another ; 

" For two reasons, therefore (viz., succession and simul 
taneity), a momentary flux is congruous and re 
mains true in regard to that which we have to 

Nor is it to be held, in acceptance of the hypothesis 
of the Vaiseshikas and Naiyayikas, that existence is a 
participation in the universal form existence ; for were 
this the case, universality, particularity, and co-inhesion 
(which do not participate in the universal) could have no 

Nor is the ascription of existence to universality, par 
ticularity, and co-inhesion dependent on any sui generis 
existence of their own ; for such an hypothesis is operose, 
requiring too many sui generis existences. Moreover, the 
existence of any universal is disproved by a dilemma 
regarding the presence or non-presence (of the one in the 
many) ; and there is not presented to us any one form 
running through all the diverse momentary things, mustard- 
seeds, mountains, and so forth, like the string running 


through the gems strung upon it. Moreover (we would 
ask), is the universal omnipresent or present everywhere in 
its subjicible subjects ? If it is everywhere, all things in 
the universe will be confounded together (chaos will be 
eternal), and you will be involved in a tenet you reject, 
since Prasasta-pcida has said, " Present in all its subjects." 
Again (if the universal is present only in its proper sub 
jects), does the universal (the nature of a jar) residing in 
an already existing jar, on being attached to another jar 
now in making, come from the one to attach itself to the 
other, or not come from it ? On the first alternative (if it 
comes), the universal must be a substance (for substances 
alone underlie qualities and motions) ; whereas, if it does 
not come, it cannot attach itself to the new jar. Again 
(we ask), when the jar ceases to exist, does the universal 
outlast it, or cease to exist, or go to another place ? On 
the first supposition it will exist without a subject to 
inhere in; on the second, it will be improper to call it 
eternal (as you do) ; on the third, it will follow that it is 
a substance (or base of qualities and motions). Destroyed 
as it is by the malign influence of these and the like 
objections, the universal is un authenticated. 

Conformably it has been said 

" Great is the dexterity of that which, existing in one 
place, engages without moving from that place in 
producing itself in another place. 

" This entity (universality) is not connected with that 
wherein it resides, and yet pervades that which 
occupies that place : great is this miracle. 

" It goes not away, nor was it there, nor is it subse 
quently divided, it quits not its former repository : 
what a series of difficulties ! " 

If you ask : On what does the assurance that the one 
exists in the many rest ? You must be satisfied with the 
reply that we concede it to repose on difference from that 
which is different (or exclusion of heterogeneity). We 
dismiss further prolixity. 


That all transmigratory existence is identical with pain 
is the common verdict of all the founders of institutes, 
else they would not be found desirous to put a stop to it 
and engaging in the method for bringing it to an end. 
We must, therefore, bear in mind that all is pain, and pain 

If you object : When it is asked, like what ? you must 
quote an instance, we reply : Not so, for momentary 
objects self-characterised being momentary, have no com 
mon characters, and therefore it is impossible to say that 
this is like that. We must therefore hold that all is like 
itself alone, like itself alone. 

In like manner we must hold that all is void, and void 
alone. For we are conscious of a determinate negation. 
This silver or the like has not been seen by me in 
sleeping or waking. If what is seen were (really) existent, 
then reality would pertain to the corresponding act of 
vision, to the (nacre, &c.), which is the basis of its par 
ticular nature (or hocceity), to the silver, &c., illusorily 
superposed upon that basis, to the connection between 
them, to the co-inherence, and so forth : a supposition not 
entertained by any disputant. Nor is a semi-effete exist 
ence admissible. No one imagines that one-half of a fowl 
may be set apart for cooking, and the other half for laying 
eggs. The venerated Buddha, then, having taught that of 
the illusorily superposed (silver, &c.), the basis (nacre, 
&c.), the connection between them, the act of vision, and 
the videns, if one or more be unreal it will perforce ensue 
that all are unreal, all being equally objects of the nega 
tion ; the Madhyamikas excellently wise explain as follows, 
viz., that the doctrine of Buddha terminates in that of a 
total void (universal baselessness or nihilism) by a slow 
progression like the intrusive steps of a mendicant, through 
the position of a momentary flux, and through the (gradual) 
negation of the illusory assurances of pleasurable sensi 
bility, of universality, and of reality. 

The ultimate principle, then, is a void emancipated from 


four alternatives, viz., from reality, from unreality, from 
both (reality and unreality), and from neither (reality nor 
unreality). To exemplify this : If real existence were the 
nature of a water-pot and the like, the activity of its 
maker (the potter) would be superfluous. 

If non-existence be its nature the same objection will 
accrue ; as it is said 

" Necessity of a cause befits not the existent, ether and 
the like, for instance ; 

" No cause is efficacious of a non-existent effect, flowers 
of the sky and the like, for instance." 

The two remaining alternatives, as self-contradictory, 
are inadmissible. It has accordingly been laid down by 
the venerated Buddha in the Alankaravatara l 

" Of things discriminated by intellect, no nature is 
ascertained ; 2 

"Those things are therefore shown to be inexplicable 

and natureless." 
And again 

" This matter perforce results, which the wise declare, 
ISTo sooner are objects thought than they are dis 

That is to say, the objects are not determined by any one 
of the four alternatives. Hence it is that it has been said 

"A religious mendicant, an amorous man, and a dog 
have three views of a woman s person, respectively that it 
is a carcass, that it is a mistress, and that it is a prey." 

In consequence, then, of these four points of view, when 
all ideas are come to an end, final extinction, which is a 
void, will result. Accordingly we have overtaken our end, 

1 Query, LanMvatara ? to which matter is reduced by the 

2 Cf. Terrier s Institutes of Meta- tactics of speculation ; and this pre- 
physic, p. 213. " If every completed dicament is described not unaptly 
object of cognition must consist of by calling it a, flux or, as we have 
object plus the subject, the object depicted it elsewhere, perhaps more 
without the subject must be incom- philosophically, as a never-ending 
plete, that is, inchoate that is, no redemption of nonsense into sense, 
possible^object of knowledge at all. and a never-ending relapse of sense 
This is the distressing predicament into nonsense." 


and there is nothing to be taught to us. There conse 
quently remain only two duties to the student interroga 
tion and acceptance. Of these, interrogation is the putting 
of questions in order to attain knowledge not yet attained. 
Acceptance is assent to the matters stated by the sacred 
teacher. These (Bauddha nihilists) are excellent in assent 
ing to that which the religious teacher enounces, and de 
fective in interrogation, whence their conventional desig 
nation of Madhyamikas (or mediocre). 

Certain other Buddhists are styled Yogacharas, because 
while they accept the four points of view proclaimed by 
the spiritual guide, and the void of external things, they 
make the interrogation : Why has a void of the internal 
(or baselessness of mental phenomena) been admitted ? 
For their technology is as follows : Self-subsistent cogni 
tion must be allowed, or it will follow that the whole 
universe is blind. It has conformably been proclaimed 
by Dharmakirti : " To one who disallows perception the 
vision of objects is not competent." 

An external percipibUe is not admissible in consequence 
of the following dilemma. Does the object cognitively 
apprehensible arise from an entity or not ? It does not 
result from an entity, for that which is generated has no 
permanence. Nor is it non-resultant, for what has not 
come into being is non-existent. Or (we may proceed) do 
you hold that a past object is cognitively apprehensible, 
as begetting cognition ? If so, this is childish nonsense, 
because it conflicts with the apparent presentness of the 
object, and because on such a supposition the sense organs 
(and other imperceptible things) might be apprehended. 
Further (we ask), Is the percipibUe a simple atom or a 
complex body ? The latter it cannot be, this alternative 
being ejected by the dilemma as to whether part or whole 
is perceived. The former alternative is equally impossible, 
an atom being supersensible, and it not being able to 
combine simultaneously with six others ; as it has been 


" If an atom could simultaneously combine with six, it 

would have six surfaces ; 
" And each of these being taken separately, there would 

be a body of atomic dimension." 

Intellect, therefore, as having no other percipibile but 
itself, is shown to be itself its own percipibile, self-sub- 
sistent, luminous with its own light, like light. Therefore 
it has been said 

"There is naught to be objectified by intellect; there is 

no cognition ulterior thereto ; 

" There being no distinction between percept and per 
cipient, intellect shines forth of itself alone." 
The identity of percipient and percept is inferrible, 
thus : That which is cognised by any cognition is not 
other than that cognition, as soul, for instance, is not other 
than the cognition of soul ; and blue and other momentary 
objects are cognised by cognitions. For if there were a 
difference (between percept and percipient), the object 
could not now have any connection with the cognition, there 
being no identity to determine a constancy of connection, 
and nothing to determine the rise of such a connection. 
As for the appearance of an interval between the object 
and subject consciousnesses, this is an illusion, like the 
appearance of two moons when there is only one. The 
cause of this illusion is ideation of difference in a stream 
without beginning and without interruption; as it has 
been said 

" As invariably cognised together, the blue object and 

the cognition thereof are identical ; 
" And the difference should be accounted for by illusory 

cognitions, as in the example of the single moon." 
And again 

" Though there is no division, the soul or intellect, by 

reason of illusory perceptions, 
" Appears to possess a duality of cognitions, of percepts 

and of percipient." 
Nor must it be supposed that (on this hypothesis) the 


juice, the energy, and the digestion derivable from an 
imaginary and an actual sweetmeat will be the same ; for 
it cannot be questioned that though the intellect be in 
strictness exempt from the modes of object and subject, 
yet there is competent to it a practical distinction in 
virtue of the succession of illusory ideas without begin 
ning, by reason of its possessing diverse modes percept 
and percipient, conformably to its illusory supposition of 
practical agency, just as to those whose eyes are dim with 
some morbid affection a hair and another minute object 
may appear either diverse or identical ; as it has been 

" As the intellect, not having object and subject modes, 
appears, by reason of illusory cognitions, 

" Illuded with the diverse forms of perception, percept 
and percipient ; 

" So when the intellect has posited a diversity, as in the 
example of the differences of the cognition of a hair 
and the like, 

" Then it is not to be doubted that it is characterised as 
percipient and percept." 

Thus it has been evinced that intellect, as affected 
by beginningless ideation, manifests itself under diverse 

When, therefore, by constancy of reflection (on the four 
points of view) aforesaid, all ideation has been interrupted, 
there arises knowledge purged from the illusions which 
take the form of objects, such illusions being now melted 
away ; and this is technically called Malwdaya (the grand 
exaltation, emancipation). 

Others again (the Sautrantikas) hold that the position 
that there is no external world is untenable, as wanting 
evidence. Nor (they contend) can it be maintained that 
invariability of simultaneous cognition is an evidence, for 
this simultaneous cognition which you accept as proof of 
the identity of subject and object is indecisive, being found 
in dubious and in contrary instances. If you rejoin (they 


proceed) : Let there be a proof of this identity, and let this 
proof be invariability of simultaneous cognition, we refuse 
this, because inasmuch as cognition must ultimately have 
some object, it is manifested in duality, and because such 
invariability of simultaneity as to time and place is im 
possible. Moreover (they continue), if the object, blue 
or whatever it be, were only a form of cognition, it 
should be presented as Ego, not as Hoc aliquid, because 
the cognition and the object would be identical. Perhaps 
you will say: A blue form consisting of cognition is 
illusorily presented as external and as other than self, and 
consequently the Ego is not suggested ; and so it has been 

" This side of knowledge which appears external to the 
other portion, 

" This appearance of duality in the unity of cognition is 

an illusion." 
And again 

" The principle to be known as internal also manifests 
itself as if it were external." 

To this we reply (say the Sautrantikas) : This is unten 
able, for if there be no external objects, there being no 
genesis of such, the comparison " as if they were external " 
is illegitimate. No man in his senses would say, " Vasu- 
mitra looks like the son of a childless mother." Again, if 
the manifestation of identity be proved by the illusoriness 
of the presentment of duality, and the presentment of 
duality be proved illusory by the manifestation of identity, 
you are involved in a logical circle. Without controversy 
we observe that cognitions take external things, blue or 
whatever they may be, as their objects, and do not take 
merely internal modifications as such, and we see that 
men in their everyday life overlook their internal states. 
Thus this argument which you adduce to prove that there 
is difference between subject and object, turns out a mere 
absurdity, like milky food made of cow-dung. When then 
you say " as if it were external," you must already suppose 


an external percipibile, and your own arrow will return 
upon you and wound you. 

If any one object that the externality of an object 
synchronous with the cognition is inadmissible, we (Sau- 
trantikas) reply that this objection is inadmissible, inasmuch 
as the subject in juxtaposition to the sensory imposes its 
form upon the cognition then in production, and the 
object is inferrible from the form thus imposed. The 
interrogation and response on this point have been thus 

"If it be asked, How can there be a past percipibile ? 
They recognise perceptibility, 

" And a competent inferribility of the individual thing 
is its imposition of its form." 

To exemplify. As nourishment is inferred from a 
thriving look, as nationality is inferred from language, 
and as affection is inferred from flurried movements, so 
from the form of knowledge a knowable may be inferred. 
Therefore it has been said 

" With half (of itself) the object moulds (the cognition) 
without losing the nature of a half ; 

" The evidence, therefore, of the recognition of a know- 
able is the nature of the knowable." 

For consciousness of the cognition cannot be the being 
of the cognition, for this consciousness is everywhere alike, 
and if indifference were to attach itself to this, it would 
reduce all things to indifference. Accordingly the formal 
argument for the existence of external things: Those things 
which while a thing exists appear only at times, all depend 
upon something else than that thing ; as, for instance, if I 
do not wish to speak or to walk, presentments of speaking 
or walking must suppose others desirous of speaking or 
walking ; and in like manner the presentments of activity 
under discussion, while there exists the recognition of a 
subject of them, are only at times manifested as blue and 
so forth. Of these, the recognition of a subject is the 
presentation of the Ego, the manifestation as blue and 


so fortli is a presentment of activity, as it has been 

" That is a recognition of a subject which is conversant 
about the Ego : 

"That is a presentment of activity which manifests 
blue and the rest." 

Over and above, therefore, the complement of subject- 
recognitions, let it be understood that there is an external 
object world perceptible, which is the cause of present 
ments of activity ; and that this external world does not 
rise into being only from time to time on occasion of pre 
sentments resulting from ideation. 

According to the view of the Sensationalists (vifnd- 
navddin), ideation is a power of generating such and 
such sensations (or presentments of activity) in subject- 
recognitions which exist as a single stream. The matur- 
escence of this power is its readiness to produce its effect ; 
of this the result is a presentment (or sensation) ; the 
antecedent momentary object (sensation) in the mental 
train is accepted as the cause, no other mental train being 
admitted to exercise such causality. It must therefore be 
stated that all momentary objects (fleeting sensations) in 
the subject-consciousness are alike able to bring about that 
maturescence of ideation in the subject-consciousness, which 
maturescence is productive of presentments of activity. 
If any one (of these fleeting sensations) had not this power, 
none would possess it, all existing alike in the stream of 
subject-recognitions. On the supposition that they all 
have this power, the effects cannot be diversified, and 
therefore any intelligent man, however unwilling, if he 
has a clear understanding, must decide, without putting 
out of sight the testimony of his consciousness, that to 
account for the occasional nature (of sense percepts) the 
six cognitions of sound, touch, colour, taste, and smell, of 
pleasure, and so forth, are produced on occasion of four 
conditions. These four conditions are known as (i.) the 
data, (2.) the suggestion, (3.) the medium, and (4.) the 


dominant (organ). Of these, the form of blue or the like 
arises from the condition of blue data in the understanding 
in which there is a manifestation of blue or the like, which 
manifestation is styled a cognition. The resuscitation of 
forms or cognitions arises from suggestion as a condition. 
The restriction to the apprehension of this or that object 
arises from the medium, light, for instance, as a condition, 
and from the dominant, the eye, for example, as another 
condition. The eye, as determinant of one particular 
cognition (form) where taste, &c., might have been equally 
cognised, is able to become dominant; for in everyday 
life he who determines is regarded as dominant. We 
must thus recognise four causes of pleasure and the rest 
which constitute the understanding and its modifications. 

So also the universe, which consists of mind and its 
modifications, is of five kinds, entitled (i.) the sensational, 
(2.) the perceptional, (3.) the affectional, (4.) the verbal, 
and (5.) the impressional. Of these, the sensible world 
(rupa-skandha) is the sense organs and their objects, 
according to the etymology, viz., that objects are discrimi 
nated (rupyante) by these. The perceptional world is the 
stream of subject-recognitions and of presentments of 
activity. The affectional world is the stream of feelings 
of pleasure and pain generated by the two aforesaid 
worlds. The verbal (or symbolical) world is the stream of 
cognitions conversant about words the words " cow," and 
so forth. The impressional world is the miseries, as desire, 
aversion, &c., caused by the affectional world, the lesser 
miseries, as conceit, pride, &c., and merit and demerit. 

Keflecting, therefore, that this universe is pain, an abode 
of pain, and an instrument of pain, a man should acquire 
a knowledge of the principles, the method of suppressing 
this pain. Hence it has been said 

" The principles sanctioned by Buddha are to the saint 
the four methods of suppressing the aggregate of 


1 Cf. Burnouf, Lotus, p. 520. Should we read samudaya ? 


In these words the sense of pain is known to every one ; 
the " aggregate " means the cause of pain. This aggregate 
is twofold, as (i.) determined by concurrence ; or (2.) deter 
mined by causation. Of these, there is an aphorism com 
prising the aggregate determined by concurrence, " which 
other causes resort to this effect ; " the condition of these 
causes thus proceeding is concurrence ; the concurrence of 
causes is the result of this only, and not of any conscious 
being, such is the meaning of the aphorism. To exemplify 
this. A germ, caused by a seed, is generated by the con 
currence of six elements. Of these, earth as an element 
produces hardness and smell in the germ; water as an 
element produces viscidity and moisture; light as an 
element produces colour and warmth ; air as an element 
produces touch and motion ; ether as an element produces 
expansion and sound ; the season as an element produces 
a fitting soil, &c. The aphorism comprising the aggregate 
determined by causation is: "With the Tathagatas the 
nature of these conditions is fixed by production, or by 
non-production ; there is continuance as a condition, and 
determination by a condition, and conformity of the pro 
duction to the cause ; " that is to say, according to the doc 
trine of the Tathagata Buddhas, the nature of these condi 
tions, that is, the causal relation between the cause and 
effect, results from production or from non-production. 
That which comes into being, provided that something 
exists, is the effect of that as its cause ; such is the expla 
nation of the nature (or causal relation). Continuance as 
a condition is where the effect is not found without its 
cause. The (abstract) affix tal (in the word sthitita) has 
the sense of the concrete. Determination by a condition 
is the determination of the effect by the cause. Here some 
one might interpose the remark that the relation of cause 
and effect cannot exist apart from some conscious agent. 
For this reason it is added that there existing a cause, 
conformity of the genesis to that cause is the nature 
which is fixed in conditions (that is, in causes and 


effects) ; and in all this no intelligent designer is observed. 1 
To illustrate this, the causal determination of a genesis to 
be gone through is as follows : From the seed the germ, 
from the germ the stalk, from the stalk the hollow stem, 
from the hollow stem the bud, from the bud the spicules, 
from the spicules the blossom, from the blossom the fruit. 
In this external aggregate neither the cause, the seed and 
the rest, nor the effect, the germ and the rest, has any 
consciousness of bringing a germ into being, or of being 
brought into being by the seed. In like manner in mental 
facts two causes are to be recognised. There is a whole 
ocean of scientific matter before us, but we desist, apprehen 
sive of making our treatise unduly prolix. 

Emancipation is the suppression of these two causal 
aggregates, or the rise of pure cognition subsequent to 
such suppression. The method (path, road) is the mode of 
suppressing them. And this method is the knowledge of 
the principles, and this knowledge accrues from former 
ideas. Such is the highest mystery. The name Sautran- 
tika arose from the fact that the venerated Buddha said 
to certain of his disciples who asked what was the ultimate 
purport (anta) of the aphorism (stitra), "As you have in 
quired the final purport of the aphorism, be Sautrantikas." 
Certain Bauddhas, though there exist the external world, 
consisting of odours, &c., and the internal, consisting of 
colours, &c., in order to produce unbelief in these, declared 
the universe to be a void. These the venerated Buddha 
styled Prathamika (primary) "disciples.. A second school, 
attached to the apprehension of sensations only, maintain 
that sensation is the only reality. A third school, who 

1 Cf G H Lewes History of property of bricks, mortar, wood, 

Philosophy, vol. i. p. 85. "We not and glass. But what we know of 

only see that the architect s plan organic materials is that they have 

determined the arrangement of this spontaneous tendency to arrange 

materials in the house, but we see themselves in definite forms ; pre- 

why it must have done so, because cisely as we see chemical substances 

the materials have no spontaneous arranging themselves in definite 

tendency to group themselves into forms without the intervention of 

houses ; that not being a recognised any extra- chemical agency." 


contend that both are true (the internal and the external), 
and maintain that sensible objects are inferrible. Others 
hold all this to be absurd language (viruddhd IMshd), and 
are known under the designation of Vaibhashikas. Their 
technical language springs up as follows: According to 
the doctrine of inferrible sensibles, there being no percep 
tible object, and consequently no object from which a 
universal rule can be attained, it will be impossible that 
any illation should take place, and therefore a contradiction 
will emerge to the consciousness of all mankind. Objects, 
therefore, are of two kinds, sensible and cogitable. Of 
these apprehension is a non- discriminative instrument of 
knowledge as other than mere representation; cognition 
which is discriminative is not a form of evidence, as being 
a merely ideal cognition. Therefore it has been said 
" Apprehension, exempt from ideality and not illusory, 
is non-discriminative. Discrimination, as resulting 
from the appearances of things, is without con 
troversy an illusion. 
"The perceptible evidence of things is perception: if 

it were aught else, 

" There could neither be things, nor evidence of things 
derived from verbal communication, inference, or 

Here some one may say : If discriminative cognition be 
unauthentic, how is the apprehension of real objects by one 
energising thereon and the universal consentiency of man 
kind to be accounted for ? Let it be replied : This question 
does not concern us, for these may be accounted for by 
the possibility of an indirect apprehension of objects, just 
as if we suppose the light of a gem to be a gem (we may 
yet handle the gem, because it underlies the light, while 
if we were to take nacre for silver, we could not lay hold 
of any silver). The rest has been fully discussed in 
describing the Sautrantikas (cf. p. 27), and therefore need 
not here be further detailed. 

It should not be contended that a diversity of instruction 



according to the disciples modes of thought is not tra 
ditional (or orthodox) ; for it is said in the gloss on the 

" The instructions of the leader of mankind (Buddha) 
accommodating themselves to the character and dis 
position (of those who are to be taught), 

" Are said to be diverse in many ways, according to a 
plurality of methods. 

" For as deep or superficial, and sometimes both deep 
and superficial, 

" Instructions are diverse, and diverse is the doctrine of 
a universal void which is a negation of duality." 

It is well known in Buddhist doctrine that the worship 
of the twelve inner seats (dyatana) is conducive to felicity. 

" After acquiring wealth in abundance, the twelve inner 

" Are to be thoroughly reverenced ; what use of reveren 
cing aught else below ? 

" The five organs of knowledge, the five organs of action, 

"The common sensory and the intellect have been 
described by the wise as the twelve inner seats." 

The system of the Buddhists is described as follows in 
the Viveka-vilasa : 

" Of the Bauddhas Sugata (Buddha) is the deity, and the 
universe is momentarily fiuxional ; 

" The following four principles in order are to be known 
by the name of the noble truths : 

" Pain, the inner seats, and from them an aggregate is 
held, 1 

" And the path (method) ; of all this let the explication 
be heard in order. 

" Pain, and the sJcandhas of the embodied one, which are 
declared to be five, 

" Sensation, consciousness, name, impression, and form. 

"The five organs of sense, the five objects of sense, 
sound and the rest, the common sensory, 

1 These are not the usual four sublime truths ; cf. p. 30. 


" And (the intellect) the abode of merit, these are the 

twelve inner seats. 
" This should be the complement of desire and so forth, 

when it arises in the heart of man. 
" Under the name of soul s own nature, it should be 

the aggregate. 

" The fixed idea that all impressions are momentary, 
" This is to be known as the path, and is also styled 

"Furthermore, there are two instruments of science, 

perception and inference. 
" The Bauddhas are well known to be divided into four 

sects, the Vaibhashikas and the rest. 
" The Vaibhashika highly esteems an object concomitant 

to the cognition ; 

"The Sautrantika allows no external object apprehen 
sible by perception ; 
"The Yogachara admits only intellect accompanied 

with forms ; 
"The Madhyamikas hold mere consciousness self-sub- 

" All the four (sects of) Bauddhas proclaim the same 

" Arising from the extirpation of desire, &c., the stream 

of cognitions and impressions. 
" The skin garment, the water-pot, the tonsure, the rags, 

the single meal in the forenoon, 
" The congregation, and the red vesture, are adopted by 

the Bauddha mendicants." x A. E. G. 

1 Mddhava probably derived most (as, e.g., that of samuddya or samu- 

of his knowledge of Buddhist doq- day a, &c.) seem to be at variance 

trines from Brahmanical works ; con- with those given in Buddhist 

sequently some of his explanations works. * 



THE Gymnosopliists l (Jainas), rejecting these opinions of 
the Muktakachchhas, 2 and maintaining continued existence 
to a certain extent, overthrow the doctrine of the momen- 
tariness of everything. (They say): If no continuing 
soul is accepted, then even the arrangement of the means 
for attaining worldly fruit in this life will be useless. 
But surely this can never be imagined as possible that 
one should act and another reap the consequences ! There 
fore as this conviction, " I who previously did the deed, 
am the person who now reap its consequences," establishes 
undoubtedly the existence of a continuing soul, which 
remains constant through the previous and the subsequent 
period, the discriminating Jaina Arhats reject as unten 
able the doctrine of momentary existence, i.e., an exist 
ence which lasts only an instant, and has no previous or 
subsequent part. 

But the opponent may maintain, " The unbroken stream 
(of momentary sensations) has been fairly proved by argu 
ment, so who can prevent it? In this way, since our 
tenet has been demonstrated by the argument, whatever 
is, is momentary, &c./ it follows that in each parallel line 
of successive experiences the previous consciousness is the 
agent and the subsequent one reaps the fruit. Nor may 

1 Virasanas, "without garments." liarity of dress, apparently a habit 

2 " The Buddhists are also called of wearing the hein of the lower 
Multakachchhas, alluding to a pecu- garment untucked." Colebrooke. 


you object that, if this were true, effects might extend 
beyond all bounds [i.e., A might act, and B receive the 
punishment] because there is an essentially controlling 
relation in the very nature of cause and effect. Thus we 
see that when mango seeds, after being steeped in sweet 
juices, are planted in prepared soil, there is a definite 
certainty that sweetness will be found in the shoot, the 
stalk, the stem, the branches, the peduncle, &c., and so on 
by an unbroken series to the fruit itself ; or again, when 
cotton seeds have been sprinkled with lac juice, there will 
be a similar certainty of finding, through the same series 
of shoot, &c., an ultimate redness in the cotton. As it 
has been said 

" In whatever series of successive states the original 
impression of the action was produced, 

" There verily accrues the result, just like the redness 
produced in cotton. 

" When lac juice, &c., are poured on the flower of the 
citron, &c., 

" A certain capacity is produced in it, do you not see 
it? " 

But all this is only a drowning man s catching at a 
straw, for it is overthrown by the following dilemma : 

In the example of the " cloud," &c. [supra, p. 15], was 
your favourite " momentariness " proved by this very proof 
or by some other ? It could not be the former, because 
your alleged momentariness is not always directly visible 
in the cloud, and consequently, as your example is not 
an ascertained fact, your supposed inference falls to the 
ground. Nor can it be the latter because you might 
always prove your doctrine of momentariness by this new 
proof (if you had it), and consequently your argument 
regarding all existence ["whatever is, is momentary," 
&c.] would become needless. If you take as your defini 
tion of " existence " " that which produces an effect," this 
will not hold, as it would include even the bite of a snake 
imagined in the rope, since this undoubtedly produces the 


effect [of fear]. Hence it has been said that the definition 
of an existence is " that which possesses an origin, an end, 
and an [intermediate] duration." 

As for what was said [in p. 16] that "the momentari- 
ness of objects is proved by the fact that the contrary 
assumption leads to contradictory attributes of capacity 
and want of capacity existing contemporaneously," that 
also is wrong for the alleged contradiction is not proved, 
as the holders of the Syad-vada 1 doctrine [vide infra] 
willingly admit the indeterminateness of the action of 
causes. As for what was said of the example of the 
cotton, that is only mere words, since no proof is given, 
and we do not accept even in that instance a separate 
destruction [at each moment]. And again, your supposed 
continued series cannot be demonstrated without some 
subject to give it coherence, as has been said, " In indi 
vidual things which are of the same class or successively 
produced or in mutual contact, there may be a continued 
series; and this series is held to be one [throughout 

Nor is our objection obviated by your supposed definite 
relation between causes and effects. For even on your 
own admission it would follow that something experienced 
by the teacher s mind might be remembered by that of 
the pupil whom he had formed, or the latter might ex 
perience the fruits of merit which the former had acquired; 
and thus we should have the twofold fault that the thing 
done passed away without result, and that the fruit of the 
thing not done was enjoyed. This has been said by the 
author of the Siddhasenavakya 

" The loss of the thing done, the enjoyment of the fruit 
of a thing not done, the dissolution of all existence, 
and the abolition of memory, bold indeed is the Buddhist 
antagonist, when, in the teeth of these four objections, 
he seeks to establish his doctrine of momentary destruc 
tion ! " 

1 In p. 26, line 3, read Sydd-vddindm. 


Moreover, (on your supposition of momentary existence), 
as at the time of the perception (the second moment) the 
object (of the first moment) does not exist, and similarly 
at the time of the object s existence the perception does 
not exist, there can be no such things as a perceiver and 
a thing perceived, and consequently the whole course of 
the world would come to an end. Nor may you suppose 
that the object and the perception are simultaneous, be 
cause this would imply that, like the two horns of an 
animal, they did not stand in the relation of cause and 
effect [as this relation necessarily involves succession], 
and consequently the Alamlana, or the object s data 
[supra, p. 29], would be abolished as one of the four con 
current causes (pratyaya). 1 

If you say that "the object may still be perceived, 
inasmuch as it will impress its form on the perception, 
even though the one may have existed in .a different 
moment from the other," this too will not hold. For if 
you maintain that the knowledge acquired by perception 
has a certain form impressed upon it, you are met by the 
impossibility of explaining how a momentary perception 
can possess the power of impressing a form ; and if you 
say that it has no form impressed upon it, you are equally 
met by the fact that, if we are to avoid incongruity, there 
must be some definite condition to determine the perception 
and knowledge in each several case. Thus by perception 
the abstract consciousness, which before existed uninflu 
enced by the external object, becomes modified under the 
form of a jar, &c., with a definite reference to each man s 
personality [i.e., I see the jar], and it is not merely the 
passive recipient of a reflection like a mirror. Moreover, 
if the perception only reproduced the form of the object, 
there would be an end of using such words as " far," 
"near," &c., of the objects. 2 Nor can you accept this 
conclusion, "as exactly in accordance with your own 

1 I propose to read in p. 26, line 5, infra, grahyasya for agrdliyasya. , 
2 As these terms necessarily relate to the perceiver. 


views," because, in spite of all our logic, the stubborn 
fact remains that we do use such phrases as " the moun 
tain is nearer " or " further," " long " or " large." NOT may 
you say that " it is the object (which supplies the form) 
that really possesses these qualities of being further, &c., 
and they are applied by a fashion of speech to the per 
ception [though not really belonging to it "] because we 
do not find that this is the case in a mirror [i.e., it does 
not become a far reflection because it represents a far 
object.] And again, as the perception produced by an 
object follows it in assuming the form of blue, so too, if 
the object be insentient, it ought equally to assume its 
form and so become itself insentient. And thus, accord 
ing to the proverb, " wishing to grow, you have destroyed 
your root," and your cause has fallen into hopeless diffi 

If, in your wish to escape this difficulty, you assert that 
" the perception does not follow the object in being in 
sentient," then there would be no perception that the 
object is insentient, 1 and so it is a case of the proverb, 
" While he looks for one thing which he has lost, another 
drops." " But what harm will it be if there is no percep 
tion of a thing s being insentient ? " [We reply], that if 
its being insentient is not perceived, while its blue form 
is perceived, the two may be quite distinct [and as different 
from each other as a jar and cloth], or it may be a case of 
" indeterminateness" [so that the two may be only occasion 
ally found together, as smoke with fire]. And again, if in- 
sentience is not perceived contemporaneously with the blue 
form, how could there then be conformity between them 
[so that both the blue and the insentience should together 
constitute the character of the thing ?] We might just as 
well maintain that, on perceiving a post, the unperceived 
universe entered into it as also constituting its character. 2 

1 I correct the reading tasydgra- may be not seen though the arayavin 
hanam to tasyd grahanam (tasyd is seen, then I may say that the post 
being jadatdydh). is the arayavin, and the unperceived 

2 /. e., if you say that the avayava three worlds its avayava / 


All this collection of topics for proof has been discussed 
at full length "by the Jaina authors, Pratapachandra and 
others, in the Prameyakamalamdrtanda, &c., and is here 
omitted for fear of swelling the book too much. 

Therefore those who wish for the summum lonum of 
man must not accept the doctrine of Buddha, but rather 
honour only the Arhata doctrine. The Arhat s nature 
has been thus described by Arhachchandra-suri, 1 in his 

" The divine Arhat is the supreme lord, the omniscient 
one, who has overcome all faults, desire, &c., adored by 
the three worlds, the declarer of things as they are." 

But may it not be objected that no such omniscient soul 
can enter the path of proof, since none of the five affirma 
tive proofs can be found to apply, as has been declared by 
Tautatita [Bhatta Kumarila 2 ] ? 

1. " No omniscient being is seen by the sense here in 
this world by ourselves or others ; nor is there any part 
of him seen which might help us as a sign to infer his 

2. "ISTor is there any injunction (vidhi) of scripture 
which reveals an eternal omniscient one, nor can the mean 
ing of the explanatory passages (arthavdda) be applied 

3. "His existence is not declared by those passages 
which refer to quite other topics ; and it cannot be con 
tained in any emphatic repetitions (anuvdda), as it had 
never been mentioned elsewhere before. 

4. "An omniscient being who had a beginning can 
never be the subject of the eternal Yeda ; and how can 
he be established by a made and spurious Veda ? 

5. "Do you say that this omniscient one is accepted on 

1 I read arhatsrarupam arkacJi- Kumarila had a little relenting to- 
chandra in p. 27, line 3, infra. wards the Jainas at the end of his life. 

2 The following passage occurs in He repented of having so cruelly per- 
some part of Kumarila s writings in secuted them, and acknowledged 
an argument against the Jainas. It that there was some truth in their 
is curious that in the Sdnkara-digvi- teaching. Jainagurumukluit kaschid 
jaya, chap. lv., it is mentioned that vidydldo jdtah. 


his own word ? How can you establish either when they 
thus both depend on reciprocal support ? 

6. " [If you say,] The saying is true because it was 
uttered by one omniscient, and this proves the Arhat s 
existence; how can either point be established without 
some previously established foundation ? 

7. "But they who accept a [supposed] omniscient on 
the baseless word of a parviscient know nothing of the 
meaning of a real omniscient s words. 


8. " And again, if we now could see anything like an 
omniscient being, we might have a chance of recognis 
ing him by the [well-known fourth] proof, comparison 

9. "And the teaching of Buddha [as well as that of Jina], 
which embraces virtue, vice, &c., would not be established 
as authoritative, if there were not in him the attribute of 
omniscience, 1 and so on." 

We reply as follows : As for the supposed contradiction 
of an Arhat s existence, derived from the failure of the 
five affirmative proofs, this is untenable, because there 
are proofs, as inference, &c., which do establish 2 his 
existence. Thus any soul will become omniscient when, 
(its natural capacity for grasping all objects remaining 
the same), the hindrances to such knowledge are done 
away. Whatever thing has a natural capacity for know 
ing any object, will, when its hindrances to such knowledge 
are done away, actually know it, just as the sense of 
vision cognises form, directly the hindrances of darkness, 
&c., are removed. Now there is such a soul, which has 
its hindrances done away, its natural capacity for grasp- 

1 Kumarila tries to prove that no would not be true and authoritative, 
such being can exist, as his existence but we see that they are, therefore 
is not established by any one of the he is omniscient." He answers by 
five recognised proofs, the sixth, retorting that the same argument 
abhdra, being negative, is, of course, might be used of Buddha by a Bud- 
not applicable. I understand the dhist; and as the Jaina himself would 
last sloka as showing the inapplic- disallow it in that case, it cannot be 
ability of " presumption " or arthd- convincing in his own. 
patti. A Jaina would say, " If the 2 In p. 29, line 2, read tatsadbhdvd- 
Arhat were not omniscient, his words vedakasya for tatsadbhdvddekasya. 


ing all things remaining unchanged; therefore there is 
an omniscient being. Nor is the assertion unestablished 
that the soul has a natural capacity for grasping all things ; 
for otherwise the Mimamsist could not maintain that a 
knowledge of all possible cases can be produced by the 
authoritative injunction of a text, 1 nor could there other 
wise be the knowledge of universal propositions, such as 
that in our favourite argument, " All things are indeter 
minate from the very fact of their existence" [and, of 
course, a follower of the Nyaya will grant that universal 
propositions can be known, though he will dispute the 
truth of this particular one], Now it is clear that the 
teachers of the Piirva Mimamsa accept the thesis that the 
soul has a natural capacity for grasping all things ; since 
they allow that a knowledge embracing all things can be 
produced by the discussion of injunctions and prohibitions, 
as is said [by Sahara in his commentary on the Sutras, 
i. i, 2], "A precept makes known the past, the present, 
the future, the minute, the obstructed, the distant, &c." 
Nor can you say that "it is impossible to destroy the 
obstructions which hinder the soul s knowing all things." 

O O * 

because we [Jainas] are convinced that there are certain 
special means to destroy these obstructions, viz., the three 
[" gems "], right intuition, &c. By this charm also, all 
inferior assaults of argument can be put to flight. 

But the Naiyayika may interpose, " You talk of the 
pure intelligence, which, after all hindrances are done 
away, sees all objects, having sense-perception at its 
height; but this is irrelevant, because there can be no 
hindrance to the omniscient, as from all eternity he has 
been always liberated." We reply that there is no proof 
of your eternally liberated being. There cannot be an 
omniscient who is eternally "liberated," from the very 
fact of his being " liberated," like other liberated persons, 
since the use of the term " liberated " necessarily im- 

1 In p. 29, line 9, for nikhildrthajnandt notpatty, I propose to read 


plies the having been previously bound ; and if the latter 
is absent, the former must be too, as is seen in the case of 
the ether. " But is not this being s existence definitely 
proved by his being the maker of that eternal series of 
effects, the earth, &c. ? according to the well-known argu 
ment, the earth, &c., must have had a maker, because they 
have the nature of effects, as a jar. " This argument, 
however, will not hold, because you cannot prove that they 
have the nature of effects. You cannot establish this from 
the fact of their being composed of parts, because this 
supposition falls upon the horns of a dilemma. Does this 
" being composed of parts " mean (i.) the being in contact 
with the parts ; or (ii.) " the being in intimate relation to 
the parts; or (iii.) the being produced from parts;" or 
(iv.) the being a substance in intimate relation ; or (v.) 
the being the object of an idea involving the notion of 
parts ? 

Not the first, because it would apply too widely, as it 
would include ether [since this, though not itself composed 
of parts, is in contact with the parts of other things ;] nor 
the second, because it would similarly include genus, &c. 
[as this resides in a substance by intimate relation, and 
yet itself is not composed of parts ;] nor the third, because 
this involves a term (" produced ") just as much disputed 
as the one directly in question ; 1 nor the fourth, because 
its neck is caught in the pillory of the following alterna 
tive : Do you mean by your phrase used above that it 
is to be a substance, and to have something else in in 
timate relation to itself, or do you mean that it must 
have intimate relation to something else, in order to 
be valid for your argument ? If you say the former, it 
will equally apply to ether, since this is a substance, and 
has its qualities resident in it by intimate relation ; if you 
say the latter, your new position involves as much dispute 
as the original point, since you would have to prove the 
existence of intimate relation in the parts, or the so-called 

1 janya is included in Kdrya and equally disputed. 


" intimate causes," which you mean by " something else." 
We use these terms in compliance with your terminology ; 
but, of course, from our point of view, we do not allow 
such a thing as " intimate relation," as there is no proof of 
its existence. 

]STor can the fifth alternative be allowed, because this 
would reach too far. as it would include soul, &c., since 
soul can be the object of an idea involving the notion 
of parts, and yet it is acknowledged to be not an effect. 1 
Nor can you maintain that the soul may still be indiscerp- 
tible in itself, but by reason of its connection with some 
thing possessing parts may itself become metaphorically 
the object of an idea involving the notion of parts, 
because there is a mutual contradiction in the idea of 
that which has no parts and that which is all-pervading, 
just as the atom [which is indiscerptible but not all- 

And, moreover, is there only one maker ? Or, again, is 
he independent ? 

In the former case your position will apply too far, as 
it will extend erroneously to palaces, &c., where we see for 
ourselves the work of many different men, as carpenters, 
&c., and [in the second case] if all the world were produced 
by this one maker, all other agents would be superfluous. 
As it has been said in the Vitardgastuti, or " Praise of 

1 . " There is one eternal maker for the world, all- 
pervading, independent, and true; they have none of 
these inextricable delusions, whose teacher art tlwu." 

And again 

2. " There is here no maker acting by his own free will, 
else his influence would extend to the making of a mat. 
What would be the use of yourself or all the artisans, if 
Iswara fabricates the three worlds ? " 

1 Tims " I- am possessed of a predicate involving the notion of 
body " (aham S ariri), " my hand," parts is applied to the soul " I." 
&c., are all sentences in which a 


Therefore it is right to hold, as we do, that omniscience 
is produced when the hindrances are removed by the three 
means before alluded to. 

Nor need the objection be made that " right intuition," 
&c., are impossible, as there is no other teacher to go to, 
because this universal knowledge can be produced by the 
inspired works of former omniscient Jinas. Nor is our 
doctrine liable to the imputation of such faults as Anyon- 
ydsrayatd?- &c., because we accept an eternal succession 
of revealed doctrines and omniscient teachers, like the end 
less series of seed springing from shoot and shoot from 
seed. So much for this preliminary discussion. 

The well-known triad called the three gems, right 
intuition, &c., are thus described in the Paramdgamasdra 
(which is devoted to the exposition of the doctrines of the 
Arhats) " Eight intuition, right knowledge, right conduct 
are the path of liberation." This has been thus explained 
by Yogadeva : 

(a.) When the meaning of the predicaments, the soul, 
&c., has been declared by an Arhat in exact accordance 
with their reality, absolute faith in the teaching, i.e., the 
entire absence of any contrary idea, is " right intuition." 
And to this effect runs the Tattvdrtha-S &tra, " Faith in the 
predicaments 2 is right intuition. " Or, as another defini 
tion gives it, " Acquiescence in the predicaments declared 
by a Jina is called right faith ; it is produced either by 
natural character or by the guru s instruction." " Natural 
character " means the soul s own nature, independent of 
another s teaching; "instruction" is the knowledge pro 
duced by the teaching of another in the form of explana 
tion, &c. 

(&.) " Eight knowledge " is a knowledge of the predica 
ments, soul, &c., according to their real nature, undisturbed 
by any illusion or doubt ; as it has been said 

1 Reasoning in a circle. I sup- that it is actually borne out in a case 

pose the &c. includes the Anavasthd- before everybody s eyes. 
dosha or reasoning ad infinitum. He 2 In p. 31, line 5, infra, read tat- 

accepts the supposed fault, and holds tvdrthe for tattvdrtham. 


" That knowledge, which embraces concisely or in detail 
the predicaments as they actually are, is called right 
knowledge by the wise." 

This knowledge is fivefold as divided into mati, sruta, 
avadhi, manas-parydya, and kevala ; as it has been said, 
"Mati, sruta, avadhi, manas-parydya, and kevala, these 
are knowledge." The meaning of this is as follows : 

1. Mati is that by which one cognises an object through 
the operation of the senses and the mind, all obstructions 
of knowledge being abolished. 

2. Sruta is the clear knowledge produced by mati, all 
the obstructions of knowledge being abolished. 

3. Avadhi is the knowledge of special objects caused 
by the abolition of hindrances, which is effected by " right 
intuition," &C. 1 

4. Manas-parydya is the clear definite knowledge of 
another s thoughts, produced by the abolition of all the 
obstructions of knowledge caused by the veil of envy. 

5. Kevala is that pure unalloyed knowledge for the sake 
of which ascetics practise various kinds of penance. 

The first of these (mati) is not self-cognised, the other 
four are. Thus it has been said - 

"True knowledge is a proof which nothing can over 
throw, and which manifests itself as well as its object ; it 
is both supersensuous and itself an object of cognition, as 
the object is determined in two ways." 

But the full account of the further minute divisions must 
be got from the authoritative treatise above-mentioned. 

(c.) " Eight conduct " is the abstaining from all actions 
tending to evil courses by one who possesses faith and 
knowledge, and who is diligent in cutting off the series of 
actions and their effects which constitutes mundane exist 
ence. This has been explained at length by the Arhat 

I . " Eight conduct is described as the entire relinquish- 

1 I read in p. 32, line 9, Samyag- by the abolition of hindrances pro- 
darsanddi for asamyagdarsanddi ; duced by the qualities, wrong in- 
but the old text may mean " caused tuition," &c. 


ment of blamable impulses ; this has been subjected to a 
fivefold division, as the five vows/ ahimsd, sunrita, asteya, 
Itralimacliaryd, and aparigraka. 1 

2. "The vow of ahimsd is the avoidance of injuring 
life by any act of thoughtlessness in any movable or 
immovable thing. 

3. " A kind, salutary, and truthful speech is called the 
vow of sunrita. That truthful speech is not truthful, 
which is unkind to others and prejudicial. 

4. "The not taking what is not given is declared to 
be the vow of asteya; the external life is a man s pro 
perty, and, when it is killed, it is killed by some one who 
seizes it. 

5. "The vow of Irahmacliaryd (chastity) is eighteen- 
fold, viz., the abandonment of all desires, 2 heavenly or 
earthly, in thought, word, and deed, and whether by one s 
own action or by one s consent, or by one s causing another 
to act. 

6. " The c vow of aparigraha is the renouncing of all 
delusive interest in everything that exists not ; since 
bewilderment of thought may arise from a delusive interest 
even in the unreal. 

7. " When carried out by the five states of mind in a 
fivefold order, these great vows of the world produce the 
eternal abode." 

The full account of the five states of mind (Ihdvand) 
has been given in the following passage [of which we only 
quote one sloka] 

" Let him carry out the vow of sunrita uninterruptedly 
by the abstinence from laughter, greed, fear, and anger, 
and by the deliberate avoidance of speech," 3 and so forth. 

These three, right intuition, right knowledge, and right 
conduct, when united, produce liberation, but not severally; 
just as, in the case of an elixir, it is the knowledge of 

1 Cf. the five yamas in the Yoga- 2 I read Jcdmdndm for Jcd?ndiuim 
sutras, 11.30. Hemachandra^6/w dA in p. 33, line 7 (2 x 3 x 3 = 18). 
81) calls them yamas. a I 1 or alhdshana, see Hemach. 16. 


what it is, faith in its virtues, and the actual application 
of the medicine, 1 united, which produce the elixir s effect, 
but not severally. 

Here we may say concisely that the tattvas or predi 
caments are two, jiva and ajiva ; the soul, jiva, is pure 
intelligence ; the non-soul, ajiva, is pure non-intelligence. 
Padmanandin has thus said 

" The two highest predicaments are soul and non- 
soul ; discrimination is the power of discriminating 
these two, in one who pursues what is to be pursued, and 
rejects what is to be rejected. The affection, &c., of the 
agent are to be rejected ; these are objects for the non- 
discriminating ; the supreme light [of knowledge] is alone 
to be pursued, which is defined as upayoga" 

Upayoga [or " the true employment of the soul s acti 
vities"] takes place when the vision of true knowledge 
recognises the manifestation of the soul s innate nature ; 
but as long] as the soul, by the bond of pradesa and the 
mutual interpenetration of form which it produces [between 
the soul and the body], considers itself as identified with 
its actions [and the body which they produce], knowledge 
should rather be defined as " the cause of its reco^nisin^ 

o o 

that it is other than these." 2 

Intelligence (chaitanyd) is common to all souls, and is 
the real nature of the soul viewed as parinata [i.e., as it is 
in itself]; but by the influence of upasamakshaya and 
kshayopaSama it appears in the "mixed" form as pos 
sessing both, 3 or again, by the influence of actions as they 
arise, it assumes the appearance of foulness, &c. 4 As has 
been said by Vachakacharya [in a sutra] 

1 I propose in p. 33, line 17, ra- 3 Or this may mean "by the in- 
sayanajndnasraddhdvachdrandni for fluence of upafama-fohaya or Tcsha- 
Tasdyatutjnanamsraddlidndvarandni. yopasama, it appears characterised 
For avachdrana, see Susruta, vol. ii. by one or the other." 

p. 157, &c. If andvarana be the 4 I read in p. 34, line 7, Tcalushd- 

true reading, I suppose it must mean dydkdrena for kalushdnydkdrena. 

"the absence of obstructions." The upasamakshaya and Tcshayopas- 

2 This is a hard passage, but some ama seem to correspond to the aupa- 
light is thrown on it by the scholiast zamika and Tcshdyika states about to 
to Hemachandra, Abtiidh, 79. be described. 



" The aupasamika, the Kshdyika, and the mixed states 
are the nature of the soul, and also the audayika and the 

1. The aupasamika state of the soul arises when all the 
effects of past actions have ceased, and no new actions 
arise [to affect the future], as when water becomes tem 
porarily pure through the defiling mud sinking to the 
bottom by the influence of the clearing nut-plant, 1 &c. 

2. The Kshdyika state arises when there is the absolute 
abolition of actions and their effects, as in final liberation. 

3. The "mixed" (misra) state combines both these, as 
when water is partly pure. 

4. The audayika state is when actions arise [exerting 
an inherent influence on the future]. The Pdrindmika 
state is the soul s innate condition, as pure intelligence, 
&c., and disregarding its apparent states, as (i), (2), (3), 
(4). 2 This nature, in one of the above-described varieties, 
is the character of every soul whether happy or unhappy. 
This is the meaning of the sutra quoted above. 

This has been explained in the Svarupa-sambodJiana 
" Not different from knowledge, and yet not identical 

w ith it, in some way both different and the same, 

knowledge is its first and last ; such is the soul described 

to be." 

If you say that, " As difference and identity are mutually 
exclusive, we must have one or the other in the case of 
the soul, and its being equally both is absurd," we reply, 
that there is no evidence to support you when you 
characterise it as absurd. Only a valid non-perception 3 
can thus preclude a suggestion as absurd ; but this is not 
found in the present case, since (in the opinion of us, the 
advocates of the Sydd-vdda) it is perfectly notorious that 
all things present a mingled nature of many contradictory 

1 StrycJiru* potatorum. 3 A valid non -perception is when 

2 Just as in the Sankhya philo- an object is not seen, and yet all the 
sophy, the soul is not really bound usual concurrent causes f visionare 
though it seems to itself to be so. present, such as the eye, light, &0. 


Others lay down a different set of tattvas from the two 
mentioned above, j iva and ajiva ; they hold that there 
are five astikdyas or categories,; -jiva, dkdsa, dharma, 
adharma, and pudgala. To all these five we can apply 
the idea of "existence" (asti), 1 as connected with the 
three divisions of time, and we can similarly apply the 
idea of " body " (kdyd)? from their occupying several parts 
of space. 

ThQjwas (souls) are divided into two, the "mundane" 
and the "released." The "mundane" pass from birth to 
birth ; and these are also divided into two, as those pos 
sessing an internal sense (samanaska), and those destitute 
of it (amanaskd). The former possesses samjnd, i.e., the 
power of apprehension, talking, acting, and receiving in 
struction ; the latter are those without this power. These 
latter are also divided into two, as " locomotive " (trasa), 
or " immovable " (sthdvard). 

The "locomotive" are those possessing at least two 
senses [touch and taste], as shell-fish, worms, &c., and are 
thus of four kinds [as possessing two, three, four, or five 
senses]; the "immovable" are earth, water, fire, air, and 
trees. 3 But here a distinction must be made. The dust 
of the road is properly " earth," but bricks, &c., are aggre 
gated " bodies of earth," and that soul by whom this body 
is appropriated becomes " earthen-bodied," and that soul 
which will hereafter appropriate it is the "earth-soul." 
The same four divisions must also be applied to the others, 
water, &c. Now the souls which have appropriated or 
will appropriate the earth, &c., as their bodies, are reckoned 
as " immovable ; " but earth, &c., and the " bodies of earth," 
&c., are not so reckoned, because they are inanimate. 4 
These other immovable things, and such as only possess 

1 I read in p. 35, line 5, stiti for kaprab hritayas trasdi chaturvidlidh 
sthiti. pritliivyaptejo. 

2 Hence the term here used for 4 In p. 35, line 1 6, I read teshdm 
"category" astikdya. ajivatvdt for teshdm jivatvdt. If we 

3 These (by Hemach. Abhidh. 21), keep the old reading we must tran- 
possess only one sense touch. In slate it, "because the former only 
p. 35> li ne 10, I read sankhagandola- are animate." 


the one sense of touch, are considered as " released," since 
they are incapable of passing into any other state of 

Dharma, adharma, and dkdsa are singular categories 
[and not generic], and they have not the attribute of 
" action," but they are the causes of a substance s change 
of place. 

Dharma, "merit," and adharma, "demerit," are well 
known. They assist souls in progressing or remaining 
stationary in the universally extended 1 sky [or ether] 
characterised by light, and also called Lokakasa; hence 
the presence of the category "merit" is to be inferred 
from progress, that of " demerit " from stationariness. The 
effect of dkdsa is seen when one thing enters into the 
space previously occupied by another. 

Pudgala, "body/ possesses touch, taste, and colour. 
Bodies are of two kinds, atomic and compound. Atoms 
cannot be enjoyed; 2 the compounds are the binary and 
other combinations. Atoms are produced by the separa 
tion of these binary and other compounds, while these 
arise from the conjunction of atoms. Compounds some 
times arise from separation and conjunction [combined] ; 
hence they are called pudgalas, because they "fill" (pur), 
and "dissolve" (gal). Although " time " is not properly 
an astikdya, because it does not occupy many separate 
parts of space [as mentioned in the definition], still it is a 
dravya [or tattva], as the definition will hold ; "substance" 
(dravya) possesses " qualities and action." 3 Qualities reside 

1 In p. 35, line 3 from bottom, I time throws himself into the Jaina 
read narratrdvasthite for sarvatrdvas- system which he is analysing, when 
thiti. In the preceding line I read we see that he gives the Jaina ter- 
dlolcendvachchhinne for dlokendvich- minology for this definition of dravya, 
clihinne. cf- Vaisesh. Sutra, i. I, 15. Par y ay a 

2 Cf. Siddhdnta-muktilvali, p. 27. is explained as barman in Hemach. 
The vishaya, is upablwga-sddhanam, Anek. Parydya, in p. 36, line II 
but it begins with the dvyanuka. This (infra, p. 53, line 9), seems used in 
category takes up the forms of sthd- a different sense from that which it 
vara which were excluded fromjira. bears elsewhere. I have taken it 

3 It is an interesting illustration doubtingly as in Hemach. Abhidh. 
how thoroughly Miidhava for the 1503, parydyo nukramah kramah 


in substance but do not themselves possess qualities, 
as the general qualities, knowledge, &c., of the jiva, form, 
&c., of the body, and the power of causing progress, 
stationariness, and motion into a place previously occu 
pied, in the case respectively of " merit," " demerit," and 
dkdsa. " Action " (parydyd) has thus been defined ; the 
actions (parydydh) of a substance are, as has been said, 
its existence, its production, its being what it is, its 
development, its course to the end, as, e.g., in the/te, the 
knowledge of objects, as of a jar, &c., happiness, pain, &c. ; 
in the pudgala, the lump of clay, the jar, &c. ; in merit 
and demerit, the special functions of progress, &c. Thus 
there are six substances or tattvas [i.e., the five above 
mentioned and " time "]. 

Others reckon the tattvas as seven, as lias been said 
" The tattvas are jiva, ajiva, dsrava, landha, samvara, 
nirjard, and moksha." Jiva and ajiva have been already 
described. Asrava is described as the movement of the 
soul called yoga, 1 through its participation in the movement 
of its various bodies, auddrika, &c. As a door opening 
into the water is called dsrava, because it causes the stream 
to descend through it, 2 so this yoga is called dsrava, be 
cause by it as by a pipe actions and their consequences 
flow in upon the soul. Or, as a wet garment collects the 
dust brought to it from every side by the wind, so the 
soul, wet with previous sins, collects, by its manifold points 
of contact with the body, the actions which are brought 
to it by yoga. Or as, when water is thrown oh a heated 
lump of iron, the iron absorbs the water altogether, so 
the jiva, heated by previous sins, receives from every side 
the actions which are brought by yoga. Kashdya (" sin," 
" defilement ") is so called because it " hurts " (kasJi) the 
soul by leading it into evil states ; it comprises anger, pride, 
delusion, and lust. Asrava is twofold, as good or evil. 
Thus abstaining from doing injury is a good yoga of the 

1 Toga seems to be here the natural 2 In line 1 8, read dsravanakdra- 
impulse of the soul to act. natvdd. 


body ; speaking what is true, measured, and profitable is a 
good yoga of the speech. 

These various subdivisions of dsrava have been described 
at length in several Sutras. " Asrava is the impulse 
to action with body, speech, or mind, and it is good or 
evil as it produces merit or demerit," &c. Others, how 
ever, explain it thus : " Asrava is the action of the senses 
which impels the soul towards external objects ; the light 
of the soul, coming in contact with external objects by 
means of the senses, becomes developed as the knowledge 
of form, &c." l 

Bandha, " bondage," is when the soul, by the influence 
of "false intuition," "non-indifference," " carelessness," and 
"sin" (Jcashdya), and also by the force of yoga, assumes 
various bodies occupying many parts of space, which enter 
into its own subtile body, and which are suited to the 
bond of its previous actions. As has been said 

"Through the influence of sin the individual soul 
assumes bodies suitable to its past actions, this is, 
bondage. " 

In this quotation the word " sin " (kashdya) is used to 
include the other three causes of bondage as well as that 
properly so termed. Vachakacharya has thus enumerated 
the causes of bondage : " The causes of bondage are false 
intuition, non-indifference, carelessness, and sin." 

(a) "False intuition" is twofold, either innate from 
one s natural character, as when one disbelieves Jaina 
doctrines from the influence of former evil actions, irre 
spectively of another s teaching, or derived, when learned 
by another s teaching. 

(&) " Non-indifference " is the non-restraint of the five 
senses, and the internal organ from the set of six, earth, 

(c) "Carelessness" (pramdda) is a want of effort to 
practise the five kinds of samiti, gupti, &c. 

1 Thejndna is one, but it becomes tion with the senses and external 
apparently manifold by its connec- objects. 


(d) " Sin " consists of anger, &c. Here we must make 
the distinction that the four things, false intuition, &c., 
cause those kinds of bondage called sthiti and anubJidva; 
yoga [or dsrava] causes those kinds called prakriti and 

" Bondage " is fourfold, as has been said : " Prakriti, 
sthiti, anubhdva, and pradesa are its four kinds." 

I. Prakriti means "the natural qualities," as bitterness 
or sweetness in the vimba plant or molasses. This may 
be subdivided into eight mtila-prakritis. 1 

Thus obstructions (dvarana) 2 cloud the knowledge and 
intuition, as a cloud obscures the sun or a shade the lamp. 
This is (a) jndndvara na, or (6) darsandvarana. (c) An object 
recognised as simultaneously existing or non-existing pro 
duces mingled pleasure and pain, as licking honey from a 
sword s edge, this is vedaniya. (d) A delusion (mohanfya) 
in intuition produces want of faith in the Jaina categories, 
like association with the wicked ; delusion in conduct pro 
duces want of self-restraint, like intoxication, (e) Ayus 
produces the bond of body, like a snare. 3 (/) Ndman, or 
" the name," produces various individual appellations, as a 
painter paints his different pictures, (g) Gotra produces 
the idea of noble and ignoble, as the potter fashions his 
pots. (A) Antardya produces obstacles to liberality, &c., 
as the treasurer hinders the king by considerations of 

Thus is t h.Qprakriti- bandha eightfold, being denominated 
as the eight mtila-pralcritis, with subdivisions according 
to the different actions of the various subject-matter. 

And thus has Umaswati-vachakacharya 4 declared: " The 
first kind of landha consists of obstructions of the know 
ledge and the intuition, vedaniya, mohaniya, dyus, ndman, 

1 These are also called the eight used for dvarana (Pan. iii. 4, 68). 
larmans in Govindananda s gloss, Cf. Yoga Sut., ii. 52, where Vya"sa s 
Ved. Sut., ii. 2, 33. Comm. has dvaraniya. 

2 The Calcutta MS. reads ddar- 3 Jdlavat ? The printed text has 
aniyasya for dvaraniyasya, in p. 37, jalavat. 

last line. But dvaraniya may be 4 Umdsvami- ? 


gotra, and antardya ; " and he has also reckoned up the 
respective subdivisions of each as five, nine, twenty-eight, 
four, two, forty, two, and fifteen. All this has been 
explained at full length in the Vidydnanda and other 
works, and here is omitted through fear of prolixity. 

2. Sthiti. As the milk of the goat, cow, buffalo, &c., 
have continued unswerving from their sweet nature for so 
long a period, so the first three mtila-prakritisjndndvarana, 
&c., and the last, antardya, have not swerved from their 
respective natures even through the period described in 
the words, " sthiti lasts beyonds crores of crores of periods 
of time measured by thirty sdgaropamas" * This con 
tinuance is sthiti. 

3. Anubh&va. As in the milk of goats, cows, buffaloes, 
&c., there exists, by its rich or poor nature, a special 
capacity for producing 2 its several effects, so in the different 
material bodies produced by our actions there exists a 
special capacity (anubhdva) for producing their respective 

4. Pradesa. The bandha called pradesa is the entrance 
into the different parts of the soul by the masses, made 
up of an endless number of parts, of the various bodies 
which are developed by the consequences of actions. 

Samvara is the stopping of dsrava that by which the 
influence of past actions (karman) is stopped from enter 
ing into the soul. It is divided into gupti, samiti, &c. 
Crupti is the withdrawal of the soul from that " impulse " 
(yoga} which causes mundane existence, it is threefold, 
as relating to body, speech, or mind. Samiti is the acting 
so as to avoid injury to all living beings. This is divided 
into five kinds, as foyd? llidslid, &c., as has been explained 
by Hemachandra. 

1 For the sdgaropama, see Wil- prachyutih sthitih for prachyutisthi- 

son s Essays, vol. i. p. 309. In till. 

p. 38, line 1 6, I read ityddyukta- 2 In p. 38, line 18, read svakdrya- 

kdldd tirdhvam api for the obscure Tcarane. 

ityddyuktam Mladurddhdnavat. I 3 In p. 39, line 2 and line 5, for 

also read at the end of the line irshyd read Iryd, a bad misreading. 


1. "In a public highway, kissed by the sun s rays, to 
walk circumspectly so as to avoid injuring living beings, 
this the good call iryd. 

2. "Let him practise 1 a measured utterance in his 
intercourse with all people ; this is called bhdshd-samiti, 
dear to the restrainers of speech. 

3. " The food which the sage takes, ever free from the 
forty-two faults which may accrue to alms, is called the 
eshand-samiti. 2 

4. " Carefully looking at it and carefully seating himself 
upon it, let him take a seat, &c., set it down, and meditate, 
this is called the dddnorsamiti. 

5. "That the good man should carefully perform his 
bodily evacuations in a spot free from all living creatures, 3 
this is the utsarga-samiti* Hence samvara has been 
etymologically analysed as that which closes (sam + vrinoti) 
the door of the stream of dsrava, 5 as has been said by the 
learned, " Asrava is the cause of mundane existence, sarp- 
vara is the cause of liberation; 6 this is the Arhat doc 
trine in a handful; all else is only the amplification of 

Nirjard is the causing the fruit of past actions to decay 
by self-mortification, &c. ; it destroys by the body the 
merit and demerit of all the previously performed actions, 
and the resulting haprfmess and misery ; " self-mortifica 
tion " means the plucking out of the hair, &c. This nir- 
jard is twofold, 7 "temporary" (yatMMla) and ancillary 
(aupakramanika). It is " temporary " as when a desire is 
dormant in consequence of the action having produced its 
fruit, and at that particular time, from this completion of 

1 In p. 39, line 6, I read dpadyetd dharma, " the ten duties of an as- 

for dpadyatd. cetic, patience, gentleness," &c. ; 

2 In p. 39, line 9, for sesliand read bhdvand, " conviction," such as that 

saishand. worldly existences are not eternal, 

3 In p. 39, line 12, join nirjantu &c.; cMritra, "virtuous observance." 
zndjagatitale. 5 In p. 39, line J 4, read dsrava- 

4 Madhava omits the remaining srotaso. 

divisions of samvara. Wilson, Essays, 6 For moha, in line 1 6, read molcsha. 
vol.i.p. 3ii,givesthemas^arwAaAd, 7 In p. 39, line 2 infra. I read 
" endurance," as of a vow ; yati- yathdkdla- for yathd kdla-. 


the object aimed at, nirjard arises, being caused by the 
consumption of the desire, &c. But when, by the force of 
asceticism, the sage turns all actions into means for attain 
ing his end (liberation), this is the nirjard of actions. 
Thus it has been said : " From the decaying of the actions 
which are the seeds of mundane existence, nirjard arises, 
which is twofold, sakdmd and akdmd. That called 
sakdmd belongs to ascetics, the akdmd to other embodied 
spirits." 1 

Mokslia. Since at the moment of its attainment there 
is an entire absence of all future actions, as all the causes 
of bondage (false perception, &c.) are stopped, 2 and since 
all past actions are abolished in the presence of the causes 
of nirjard, there arises the absolute release from all actions, 
this is moksha ; as it has been said: "Mokslia is the 
absolute release from all actions by the decay (nirjard} of 
the causes of bondage and of existence." 

Then the soul rises upward to the end of the world. 
As a potter s wheel, whirled by the stick and hands, moves 
on even after these have stopped, until the impulse is 
exhausted, so the previous repeated contemplations of the 
embodied soul for the attainment of mdksha exert their influ 
ence even after they have ceased, and bear the soul onward 
to the end of the world ; or, as the gourd, encased with 
clay, sinks in the water, but rises to the surface when freed 
from its encumbrance, so the soul, delivered from works, 
rises upward by its isolation, 3 from the bursting of its 
bonds like the elastic seed of the castor-oil plant, or by its 
own native tendency like the flame. 

1 This passage is very difficult and dormant ; the latter is salcdmd, be- 
not improbably corrupt, and my in- cause the ascetic conquers the lower 
terpretation of it is only conjectural, desire under the overpowering influ- 
The ordinary nirjard is when an ence of the higher desire for libera- 
action attains its end (like the lull- tion. 

ing of a passion by the gratification), 2 I read nirodhe for nirodhah in 

this lull is temporary. That nirjard p. 40, line 6 ; cf. p. 37, line 13. The 

is "ancillary" which is rendered by causes of bondage produce the as- 

asceticism a means to the attainment sumption of bodies in which future 

of the highest good. The former is actions are to be performed. 

aJcdmd, " desireless," because at the 3 Literally " absence of sanga. 
moment the desire is satisfied and so 


" Bondage " is the condition of being unseparated, with 
a mutual interpenetration of parts [between the soul and 
the body] ; sanga is merely mutual contact. This has 
been declared as follows : 

"[Liberation] is unhindered, from the continuance of 
former impulses, from the absence of sanga, from the cut 
ting of all bonds, and from the natural development of the 
soul s own powers of motion, like the potter s wheel, the 
gourd with its clay removed, the seed of the castor-oil 
plant, or the flame of fire." 

Hence they recite a sloka : 

" However often they go away, the planets return, the 
sun, moon, and the rest ; 

"But never to this day have returned any who have 
gone to Alokakasa." 

Others hold moksha .to be the abiding in the highest 
regions, the soul being absorbed in bliss, with its know 
ledge unhindered and itself untainted by any pain or im 
pression thereof. 

Others hold nine tattwas, adding "merit" and "demerit" 
to the foregoing seven, these two being the causes of 
pleasure and pain. This has been declared in the Sid- 
dhdnta, " Jiva, ajiva, puny a, pdpa, dsrava, samvara, nir- 
jarana, landlia, and mokslia, are the nine tattiuas" As 
our object is only a summary, we desist here. 

Here the Jainas everywhere introduce their favourite 
logic called the sapta-lliangi-naya^ or the system of the 
seven paralogisms, " may be, it is," " may be, it is not," 
" may be, it is and it is not," " may be, it is not predicable," 
" may be, it is, and yet not predicable," " may be, it is not, 
and not predicable," " may be, it is and it is not, and not 
predicable." All this Anantavirya has thus laid down : 

1. "When you wish to establish a thing, the proper 
course is to say may be, it is ; when you wish to deny 
it, may be, it is not/ 

2. " When you desire to establish each in turn, let your 

1 In p. 41, line 7, read saptalhanginaya, see Ved. S. Gloss., ii. 2, 23. 


procedure likewise embrace both ; when you wish to 
establish both at once, let it be declared indescribable 
from the impossibility to describe it. 

3. "The fifth process is enjoined when you wish to 
establish the first as well as its indescribableness ; when 
the second as well as its indescribableness, the occasion 
for the sixth process arises. 

4. " The seventh is required when all three characters 
are to be employed simultaneously." 

Sydt, " may be," is here an indeclinable particle in the 
form of a part of a verb, used to convey the idea of in- 
determinateness ; as it has been said 

" This particle sydt is in the form of a verb, but, from 
its being connected with the sense, it denotes 
indeterminateness in sentences, and has a qualify 
ing effect on the implied meaning." 

If, again, the word sydt denoted determinateness, then 
it would be needless in the phrase, " may be, it is ; " but 
since it really denotes indeterminateness, " may be, it is," 
means "it is somehow;" sydt, "may be," conveys the 
meaning of " somehow," kathamchit ; and so it is not 
really useless. As one has said 

" The doctrine of the sydd-vdda arises from our every 
where rejecting the idea of the absolute ; 1 it depends on 
the sapta-lhangi-naya, and it lays down the distinction 
between what is to be avoided and to be accepted." 

If a thing absolutely exists, it exists altogether, always, 
everywhere, and with everybody, and no one at any time or 
place would ever make an effort to obtain or avoid it, as 
it would be absurd to treat what is already present as an 
object to be obtained or avoided. But if it be relative (or 
indefinite), the wise will concede that at certain times and 
in certain places any one may seek or avoid it. More 
over, suppose that the question to be asked is this : " Is 
being or non-being the real nature of the thing ? " The 

1 I cannot understand the words tadvidheh, and therefore leave them 
at the end of the first line, kim vrita- untranslated. 


real nature of the thing cannot be being, for then you 
could not properly use the phrase, " It is a pot " (ghatosti), 
as the two words " is " and " pot " would be tautological ; 
nor ought you to say, " It is not a pot," as the words thus 
used would imply a direct contradiction ; and the same 
argument is to be used in other questions. 1 As it has 
been declared 

" It must not be said It is a pot, since the word pot 
implies is ; 

" E"or may you say it is not a pot, for existence and 
non-existence are mutually exclusive," &c. 

The whole is thus to be summed up. Four classes of 
our opponents severally hold the doctrine of existence, 
non-existence, existence and non-existence successively, 
and the doctrine that everything is inexplicable (anirva- 
chaniyatd) ; 2 three other classes hold one or other of the 
three first theories combined with the fourth. 3 Now, when 
they meet us with the scornful questions, " Does the thing 
exist ? " &c., we have an answer always possible, " It exists 
in a certain way," &c., and our opponents are all abashed 
to silence, and victory accrues to the holder of the Sydd- 
vdda, which ascertains the entire meaning of all things. 
Thus said the teacher in the Syddvdda-manjari 

"A thing of an entirely indeterminate nature is the 
object only of the omniscient ; a thing partly determined 
is held to be the true object of scientific investigation. 4 
When our reasonings based on one point proceed in the 
revealed way, it is called the revealed Sydd-vdda, which 
ascertains the entire meaning of all things." 

" All other systems are full of jealousy from their mutual 
propositions and counter-propositions ; it is only the doc 
trine of the Arhat which with no partiality equally favours 
all sects." 

1 Thus Govindananda applies it tenet in the Kliandana-Tchanda-Wid- 
( Ved. 8 tit., ii. 2, 33) to " may be dya. 

it is one," "may be it is many," 3 In p. 42, line 1 7, for matendmisri- 
&c. tdni read matena mUritdni. 

a AjeaTaXi^ia. This is Sriharsha s 4 In p. 43, line 2, for na yasya 

read nayasya. 


The Jaina doctrine has thus been summed up by 

" The hindrances belonging to vigour, enjoyment, sensual 
pleasure, giving and receiving, sleep, fea,r, ignorance, aver 
sion, laughter, liking, disliking, love, hatred, want of in 
difference, desire, sorrow, deceit, these are the eighteen 
faults (dosha) according to our system. 1 The divine 
Jina is our G-uru, who declares the true knowledge of the 
tattwas. The path 2 of emancipation consists of knowledge, 
intuition, and conduct. There are two means of proof 
(pramdna) in the Sydd-vdda doctrine, sense-perception 
and inference. All consists of the eternal and the non- 
eternal ; there are nine or seven tattwas. The jiva t the 
ajiva, merit and demerit, dsrava, samvara, landha, nirjard, 
mukti, we will now explain each. Jiva is denned as 
intelligence ; ajiva is all other than it ; merit means bodies 
which arise from good actions, demerit the opposite ; 
dsrava is the bondage of actions, 3 nirjard is the unloosing 
thereof ; moksha arises from the destruction of the eight 
forms of karman or "action." But by some teachers 
" merit " is included in samvara* and " demerit " in dsrava. 

" Of the soul which has attained the four infinite things 5 
and is hidden from the world, and whose eight actions are 
abolished, absolute liberation is declared by Jina. The 
Swetambaras are the destroyers of all defilement, they 
live by alms, 6 they pluck out their hair, they practise 
patience, they avoid all association, and are called the 
Jaina Sddhus. The Digambaras pluck out their hair, they 

1 This list is badly printed in the 3 This seems corrupt, a line is 
Calcutta edition. It is really identi- probably lost. 

cal with that given in Hemachandra s 4 In last line, for samsrare read 

Abhidhdna-chintdmani, 72, 73; but samvare. 

we must correct the readings to 6 Does this mean the knowledge 

antardyds, rdgadweslidv ariratih sma- of the world, the soul, the liberated 

rah, and Jidso for himsd. The order and liberation ? These are called 

of the eighteen doshas in the Cal- ananta. See Weber s Bhagavati, 

cutta edition is given by Hema- pp. 250, 261-266. 

chandra as 4, 5, i, 2, 3, 10, 1 1, 12, 6 Sarajohara-ridh is explained by 

7, 9, 17, 1 6, 1 8, 8, 6, 15, 13, 14. the rajoharanadhdrin (= vratin} of 

2 In p. 43, line 13, for vartini read Halayudha, ii. 189. 


carry peacocks tails in their hands, they drink from their 
hands, and they eat upright in the giver s house, these 
are the second class of the Jaina Rishis. 

"A woman attains not the highest knowledge, she 
enters not Mukti, so say the Digambaras ; but there is 
a great division on this point between them and the 
Swetambaras. 1 E. B. C. 

1 Cf. Wilson, Essays, i. 340. For strim read stri. 




Tins doctrine of the Arhatas deserves a rational con 
demnation, for whereas there is only one thing really 
existent, the simultaneous co-existence of existence, non- 
existence and other modes in a plurality of really existing 
things is an impossibility. Nor should any one say : 
Granting the impossibility of the co-existence of exist 
ence and non-existence, which are reciprocally contra 
dictory, why should there not be an alternation between 
existence and non-existence? there being the rule that 
it is action, not Ens, that alternates. Nor let it be sup 
posed that the whole universe is multiform, in reliance 
upon the examples of the elephant-headed Ganesa and of 
the incarnation of Vishnu as half man, half lion ; for 
the elephantine and the leonine nature existing in one 
part, and the human in another, and consequently there 
being no contradiction, those parts being different, these 
examples are inapplicable to the maintenance of a nature 
multiform as both existent and non-existent in one and 
the same part (or place). Again, if any one urge : Let 
there be existence in one form, and non-existence in 
another, and thus both will be compatible ; we rejoin : 
Not so, for if you had said that at different times existence 
and non-existence may be the nature of anything, then 
indeed there would have been no vice in your procedure. 
Nor is it to be contended : Let the multiformity of the 
universe be like the length and shortness which pertain 


to the same thing (in different relations) ; for in these (in 
this length and shortness) there is no contrariety, in 
asmuch as they are contrasted with different objects. 
Therefore, for want of evidence, existence and non-exist 
ence as reciprocally contradictory cannot reside at the 
same time in the same thing. In a like manner may be 
understood the refutation of the other Wiangas (Arhata 

Again, we ask, is this doctrine of the seven Wiangas t 
which lies at the base of all this, itself uniform (as ex 
cluding one contradictory), or multiform (as conciliating 
contradictories). If it is uniform, there will emerge a 
contradiction to your thesis that all things are multiform ; 
if it is multiform, you have not proved what you wished 
to prove, a multiform statement (as both existent and 
non-existent) proving nothing. 1 In either case, there is 
rope for a noose for the neck of the Syad-Vadin. 

An admirable author of institutes has the founder of 
the Arhata system, dear to the gods (uninquiring pietist), 
proved himself to be, when he has not ascertained whether 
his result is the settling of nine or of seven principles, 
nor the investigator who settles them, nor his organon, the 
modes of evidence, nor the matter to be evidenced, whether 
it be ninefold or not ! 

In like manner if it be admitted that the soul has (as 
the Arhatas say), an extension equal to that of the body, 
it will follow that in the case of the souls of ascetics, who 
by the efficacy of asceticism assume a plurality of bodies, 

1 Cf . " The argument in defence Herakleitean must go through like 
of the Maxim of Contradiction is other persons, and when, if he pro- 
that it is a postulate employed in ceeded upon his own theory, he could 
all the particular statements as to neither give nor receive information 
matters of daily experience that a by speech, nor ground any action 
man understands and acts upon when upon the beliefs which he declares 
heard from his neighbours ; a postu- to co-exist in his own mind. Ac- 
late such that, if you deny it, no cordingly the Herakleitean Kratylus 
speech is either significant or trust- (so Aristotle says) renounced the 
worthy to inform and guide those use of affirmative speech, and simply 
who hear it. You may cite innu- pointed with his finger." Grote s 
merable examples both of speech and Aristotle, vol. ii. pp. 297, 298. 
action in the detail of life, which the 


there is a differentiation of the soul for each of those bodies. 
A soul of the size of a human body would not (in the 
course of its transmigrations) be able to occupy the whole 
body of an elephant; and again, when it laid aside its 
elephantine body to enter into that of an ant, it would lose 
its capacity of filling its former frame. And it cannot be 
supposed that the soul resides successively in the human, 
elephantine, and other bodies, like the light of a lamp 
which is capable of contraction and expansion, according 
as it occupies the interior of a little station on the road 
side in which travellers are supplied with water, or the 
interior of a stately mansion ; for it would follow (from 
such a supposition) that the soul being susceptible of 
modifications and consequently non-eternal, there would 
be a loss of merits and a fruition of good and evil un 

As if then we had thrown their best wrestler, the re- 
dargution of the rest of their categories may be anticipated 
from this exposition of the manner in which their treat 
ment of the soul has been vitiated. 

Their doctrine, therefore, as repugnant to the eternal, 
infallible revelation, cannot be adopted. The venerated 
Vyasa accordingly propounded the aphorism (ii. 2, 33), 
" Nay, because it is impossible in one ; " and this same 
aphorism has been analysed by Eamanuja with the ex 
press purpose of shutting out the doctrine of the Jainas. 
The tenets of Eamanuja are as follows : Three categories 
are established, as soul, not-soul, and Lord; or as sub 
ject, object, and supreme disposer. Thus it has been 

"Lord, soul, and not-soul are the triad of principles: 
Hari (Vishnu) 

" Is Lord ; individual spirits are souls ; and the visible 
world is not-soul." 

Others, again (the followers of Sankaracharya), maintain 
that pure intelligence, exempt from all differences, the 
absolute, alone is really existent ; and that this absolute 


whose essence is eternal, pure, intelligent, and free, the 
identity of which with the individuated spirit is learnt 
from the "reference to the same object" (predication), 
" That art thou," undergoes bondage and emancipation. 
The universe of differences (or conditions) such as that of 
subject and object, is all illusorily imagined by illusion as 
in that (one reality), as is attested by a number of texts : 
Existent only, fair sir, was this in the beginning, One only 
without a second, and so forth. Maintaining this, and 
acknowledging a suppression of this beginningless illusion 
by knowledge of the unity (and identity) of individuated 
spirits and the undifferenced absolute, in conformity with 
hundreds of texts from the Upanishads, such as He that 
knows spirit passes beyond sorrow ; rejecting also any 
real plurality of things, in conformity with the text con 
demnatory of duality, viz., Death after death he undergoes 
who looks upon this as manifold ; and thinking themselves 
very wise, the Sankaras will not tolerate this division 
(viz., the distribution of things into soul, not-soul, and 
Lord). To all this the following counterposition is laid 
down : This might be all well enough if there were any 
proof of such illusion. But there is no such ignorance (or 
illusion), an unbeginning entity, suppressible by know 
ledge, testified in the perceptions, I am ignorant, I know 
not myself and other things. Thus it has been said (to 
explain the views of the Sankara) 

"Entitative from everlasting, which is dissolved by 

" Such is illusion. This definition the wise enunciate." 

This perception (they would further contend) is not 
conversant about the absence of knowledge. Eor who 
can maintain this, and to whom ? One who leans on the 
arm of Prabhakara, or one to whom Kumarila-bhatta gives 
his hand ? Not the former, for in the words 

" By means of its own and of another s form, eternal in 
the existent and non-existent, 

" Thing is recognised something by some at certain times. 


" Non-entity is but another entity by some kind of 
relation. Non-entity is but another entity, naught 
else, for naught else is observed." 

They deny any non-entity ulterior to entity. Non 
entity being cognisable by the sixth instrument of know 
ledge (anupaldbdhi), and knowledge being always an object 
of inference, the absence of knowledge cannot be an object 
of perception. If, again, any one who maintains non-entity 
to be perceptible should employ the above argument (from 
the perceptions, I am ignorant, I know not myself, and 
other things) ; it may be replied : " Is there, or is there 
not, in the consciousness, I am ignorant, an apprehension 
of self as characterised by an absence, and of knowledge 
as the thing absent or non-existent ? If there is such 
apprehension, consciousness of the absence of knowledge 
will be impossible, as involving a contradiction. If there 
is not, consciousness of the absence of knowledge, which 
consciousness presupposes a knowledge of the subject and 
of the thing absent, will not readily become possible. In 
asmuch (the Sankaras continue) as the foregoing difficul 
ties do not occur if ignorance (or illusion) be entitative, 
this consciousness (I am ignorant, I know not myself, and 
other things) must be admitted to be conversant about an 
entitative ignorance. 

All this (the Eamanuja replies) is about as profitable as 
it would be for a ruminant animal to ruminate upon ether ; 
for an entitative ignorance is not more supposable than 
an absence of knowledge. For (we would ask), is any 
self-conscious principle presented as an object and as a 
subject (of ignorance) as distinct from cognition ? If it is 
presented, how, since ignorance of a thing is terminable by 
knowledge of its essence, can the ignorance continue ? If 
none such is presented, how can we be conscious of an 
ignorance which has no subject and no object ? If you say: 
A pure manifestation of the spiritual essence is revealed 
only by the cognition opposed to ignorance (or illusion), 
and thus there is no absurdity in the consciousness of ignor- 


ance accompanied with a consciousness of its subject 
and object ; then we rejoin : Unfortunately for you, this 
(consciousness of subject) must arise equally in the absence 
of knowledge (for such we define illusion to be), notwith 
standing your assertion to the contrary. It must, there 
fore, be acknowledged that the cognition, I am ignorant, 
I know not myself and other things, is conversant about 
an absence of cognition allowed by us both. 

Well, then (the Saiikaras may contend), let the form of 
cognition evidentiary of illusion, which is under disputa 
tion, be inference, as follows : Eight knowledge must have 
had for its antecedent another entity (sc. illusion), an entity 
different from mere prior non-existence of knowledge, 
which envelops the objects of knowledge, which is ter 
minable by knowledge, which occupies the place of know 
ledge, inasmuch as it (the right knowledge) illuminates an 
object not before illuminated, like the light of a lamp 
springing up for the first time in the darkness. This argu 
ment (we reply) will not stand grinding (in the dialectic 
mill) ; for to prove the (antecedent) illusion, you will 
require an ulterior illusion which you do not admit, and a 
violation of your own tenets will ensue, while if you do 
not so prove it, it may or may not exist ; and, moreover, 
the example is incompatible with the argument, for it can 
not be the lamp that illumines the hitherto unillumined 
object, since it is knowledge only that illumines ; and an 
illumination of objects may be effected by knowledge 
even without the lamp, while the light of the lamp is only 
ancillary to the visual organ which effectuates the cogni 
tion, ancillary mediately through the dispulsion of the 
obstruent darkness. We dismiss further prolixity. 

The counterposition (of the Eamanujas) is as follows : 
The illusion under dispute does not reside in Brahman, 
who is pure knowledge, because it is an illusion, like the 
illusion about nacre, &c. If any one ask: Has not the 
self-conscious entity that underlies the illusion about 
nacre, &c., knowledge only for its nature ? they reply : 


Do not start such difficulties ; for we suppose that con 
sciousness by its bare existence has the nature of creating 
conformity to the usage about (i.e., the name and notion 
of) some object ; and such consciousness, also called know 
ledge, apprehension, comprehension, intelligence, &c., con 
stitutes the soul, or knowledge, of that which acts and 
knows. If any one ask: How can the soul, if it con 
sists of cognition, have cognition as a quality? they 
reply: This question is futile; for as a gem, the sun, 
and other luminous things, existing in the form of light, 
are substances in which light as a quality inheres for 
light, as existing elsewhere than in its usual receptacle, 
and as being a mode of things though a substance, is still 
styled and accounted a quality derived from determination 
by that substance, so this soul, while it exists as a. self- 
luminous intelligence, has also intelligence as its quality. 
Accordingly the Vedic texts : A lump of salt is always 
within and without one entire mass of taste, so also this 
soul is within and without an entire mass of knowledge ; 
Herein this person is itself a light ; Of the knowledge of 
that which knows there is no suspension ; He who knows, 
smells this ; and so also, This is the soul which, consisting 
of knowledge, is the light within the heart ; For this per 
son is the seer, the hearer, the taster, the smeller, the 
thinker, the understander, the doer ; The person is know 
ledge, and the like texts. 

It is not to be supposed that the Veda also affords 
evidence of the existence of the cosmical illusion, in the 
text, Enveloped in untruth (anrita) ; for the word untruth 
(anrita) denotes that which is other than truth (rita). 
The word rita has a passive sense, as appears from the 
words, Drinking rita. Rita means works done without 
desire of fruit ; having as its reward the attainment of the 
bliss of the Supreme Spirit through his propitiation. In 
the text in question, untruth (anrita) designates the scanty 
fruit enjoyed during transmigratory existence as opposed to 
that (which results from propitiation of the Supreme Spirit), 


which temporal fruit is obstructive to the attainment of 
supreme existence (brahman) ; the entire text (when the 
context is supplied) being : They who find not this sup 
reme sphere (brahma-loka) are enveloped in untruth. In 
such texts, again, as Let him know illusion (mdyd) to be 
the primary emanative cause (praJcriti), the term (mdyd) 
designates the emanative cause, consisting of the three 
" cords " (guna), and creative of the diversified universe. 
It does not designate the inexplicable illusion (for which 
the Sankaras contend). 

In such passages as, By him the defender of the body of 
the child, moving rapidly, the thousand illusions (mdyd) of 
the barbarian were swooped upon as by a hawk, we observe 
that the word "illusion" (mdyd) designates the really 
existent weapon of a Titan, capable of projective diversified, 
creation. The Veda, then, never sets out an inexplicable 
illusion. Nor (is the cosmical illusion to be inferred from 
the "grand text," That art thou), inasmuch as the words, 
That art thou, being incompetent to teach unity, and in 
dicating a conditionate Supreme Spirit, we cannot under 
stand by them the essential unity of the mutually exclusive 
supreme and individual spirits ; for such a supposition (as 
that they are identical) would violate the law of excluded 
middle. To explain this. The term That denotes the 
Supreme Spirit exempt from all imperfections, of illimit 
able excellence, a repository of innumerable auspicious 
attributes, to whom the emanation, sustentation, retracta 
tion of the universe is a pastime ; x such being the Supreme 
Spirit, spoken of in such texts as, That desired, let me be 
many, let me bring forth. Perhaps the word Thou, refer 
ring to the same object (as the word That), denotes the 
Supreme Spirit characterised by consciousness, having all 
individual spirits as his body; for a "reference to the 
same object" designates one thing determined by two 
modes. Here, perhaps, an Advaita-vadin may reply : Why 

i Of. the dictum of Herakleitus : p. 803) : Man is made to be the 
Making worlds is Zeus s pastime ; plaything of God. 
and that of Plato (Laws, Book vii. 


may not the purport of the reference to the same object 
in the words, That art thou, be undifferenced essence, the 
unity of souls, these words (That and thou) having a 
(reciprocally) implicate power by abandonment of opposite 
portions of their meaning ; as is the case in the phrase, 
This is that Devadatta. In the words, This is that Deva- 
datta, we understand by the word That, a person in rela 
tion to a different time and place, and by the word This, 
a person in relation to the present time and place. That 
both are one and the same is understood by the form of 
predication ("reference to the same object"). Now as 
one and the same thing cannot at the same time be known 
as in different times and places, the two words (This and 
That) must refer to the essence (and not to the accidents 
of time and place), and unity of essence can be understood. 
Similarly in the text, That art thou, there is implicated 
an indivisible essence by abandonment of the contradictory 
portions (of the denotation), viz., finite cognition (which 
belongs to the individual soul or Thou), and infinite cog 
nition (which belongs to the real or unindividual soul). 
This suggestion (the Bamanujas reply) is unsatisfactory, 
for there is no opposition (between This and That) in the 
example (This is that Deva-datta), and consequently not 
the smallest particle of " implication " (lakshand, both This 
and That being used in their denotative capacity). The 
connection of one object with two times past and present 
involves no contradiction. And any contradiction sup 
posed to arise from relation to different places may be 
avoided by a supposed difference of time, the existence in 
the distant place being past, and the existence in the near 
being present. Even if we concede to you the "implica 
tion," the (supposed) contradiction being avoidable by sup 
posing one term (either That or Thou) to be implicative, it 
is unnecessary to admit that both words are implicative. 
Otherwise (if we admit that both words are implicative), 
if it be granted that the one thing may be recognised, 
with the concomitant assurance that it differs as this and 


as that, permanence in things will be inadmissible, and 
the Buddhist assertor of a momentary flux of things will 
be triumphant. 

We have, therefore (the Bamanujas continue), laid it 
down in this question that there is no contradiction in the 
identity of the individual and the Supreme Spirit, the 
individual spirits being the body and the Supreme Spirit 
the soul. For the individual spirit as the body, and there 
fore a form, of the Supreme Spirit, is identical with the 
Supreme Spirit, according to another text, Who abiding 
in the soul, is the controller of the soul, who knows the 
soul, of whom soul is the body. 

Your statement of the matter, therefore, is too narrow. 
ALL words are designatory of the Supreme Spirit. They 
are not all synonymous, a variety of media being possible ; 
thus as all organised bodies, divine, human, &c., are forms 
of individual spirits, so all things (are the body of Sup 
reme Spirit), all things are identical with Supreme Spirit. 

God, Man, Yaksha, Pisacha, serpent, Eakshasa, bird, 
tree, creeper, wood, stone, grass, jar, cloth, these and all 
other words, be they what they may, which are current 
among mankind as denotative by means of their base and 
its suffixes, as denoting those things, in denoting things of 
this or that apparent constitution, really denote the in 
dividual souls which assumed to them such body, and the 
whole complexus of things terminating in the Supreme 
Spirit ruling within. That God and all other words what 
soever ultimately denote the Supreme Spirit is stated in 
the Tattvamuktavali and in the Chaturantara 

" God, and all other words, designate the soul, none else 

than That, called the established entity, 
" Of this there is much significant and undoubted 

exemplification in common speech and in the 

"Existence when dissociated from spirit is unknown; 

in the form of gods, mortals, and the rest 


" When pervading the individual spirit, the infinite 
has made a diversity of names and forms in the 

In these words the author, setting forth that all words, 
God, and the rest, designate the body, and showing in the 
words, " No unity in systems," &c., the characteristic of 
body, and showing in the words, " By words which are sub 
stitutes for the essence of things," &c., that it is established 
that nothing is different from the universal Lord, lays down 
in the verses, Significant of the essence, &c., that all words 
ultimately designate the Supreme Spirit. All this may be 
ascertained from that work. The same matter has been 
enforced by Bamanuja in the Vedartha-sangraha, when 
analysing the Vedic text about names and forms. 

Moreover, every form of evidence having some deter 
minate object, there can be no evidence of an undetermined 
(unconditionate) reality. Even in non-discriminative per 
ception it is a determinate (or conditioned) thing that is 
cognised. Else in discriminative perception there could 
not be shown to be a cognition characterised by an already 
presented form. Again, that text, That art thou, is not 
sublative of the universe as rooted in illusion, like a sen 
tence declaratory that what was illusorily presented, as a 
snake is a piece of rope ; nor does knowledge of the unity 
of the absolute and the soul bring (this illusory universe) 
to an end ; for we have already demonstrated that there 
is no proof of these positions. 

JSTor is there an absurdity (as the Sankaras would say), 
on the hypothesis enunciatory of the reality of the universe, 
in affirming that by a cognition of one there is a cognition 
of all things : for it is easily evinced that the mundane 
egg, consisting of the primary cause (prakriti), intellect, 
self-position, the rudimentary elements, the gross elements, 
the organs (of sense and of action), and the fourteen worlds, 
and the gods, animals, men, immovable things, and so 
forth, that exist within it, constituting a complex of all 
forms, is all an effect, and that from the single cognition 


of absolute spirit as its (emanative) cause, when we recog 
nise that all this is absolute spirit (there being a tautology 
between cause and effect), there arises cognition of all 
things, and thus by cognition of one cognition of all. Be 
sides, if all else than absolute spirit were unreal, then all 
being non-existent, it would follow that by one cognition 
all cognition would be sublated. 

It is laid down (by the Kamanujas) that retractation 
into the universe (pralaya) is when the universe, the body 
whereof consists of souls and the originant (prakriti), 
returns to its imperceptible state, unsusceptible of division 
by names and forms, existing as absolute spirit the emana 
tive cause ; and that creation (or emanation) is the gross 
or perceptible condition of absolute spirit, the body whereof 
is soul and not soul divided by diversity of names and 
forms, in the condition of the (emanative) effect of absolute 
spirit. In this way the identity of cause and effect laid 
down in the aphorism (of Vyasa) treating of origination, 
is easily explicable. The statements that the Supreme 
Spirit is void of attributes, are intended (it is shown) to 
deny thereof phenomenal qualities which are to be escaped 
from by those that desire emancipation. The texts which 
deny plurality are explained as allowed to be employed 
for the denial of the real existence of things apart from 
the Supreme Spirit, which is identical with all things, it 
being Supreme Spirit which subsists under all forms as 
the soul of all, all things sentient and unsentient being 
forms as being the body of absolute Spirit. 1 

What is the principle here involved, pluralism or monism, 
or a universe both one and more than one? Of these 
alternatives monism is admitted in saying that Supreme 
Spirit alone subsists in all forms as all is its body ; both 
unity and plurality are admitted in saying that one only 
Supreme Spirit subsists under a plurality of forms diverse 
as soul and not-soul ; and plurality is admitted in saying 

1 "Whose body nature is, and God the soul." Pope. 


that the essential natures of soul, not-soul, and the Lord, 
are different, and not to be confounded. 

Of these (soul, not-soul, and the Lord), individual 
spirits, or souls, consisting of uncontracted and unlimited 
pure knowledge, but enveloped in illusion, that is, in 
works from all eternity, undergo contraction and expan 
sion of knowledge according to the degrees of their merits. 
Soul experiences fruition, and after reaping pleasures and 
pains proportionate to merits and demerits, there ensues 
knowledge of the Lord, or attainment of the sphere of the 
Lord. Of things which are not-soul, and which are objects 
of fruition (or experience of pleasure and pain), uncon 
sciousness, unconduciveness to the end of man, suscepti 
bility of modification, and the like, are the properties. 
Of the Supreme Lord the attributes are subsistence, as 
the internal controller (or animator) of both the subjects 
and the objects of fruition ; the boundless glory of illimi 
table knowledge, dominion, majesty, power, brightness, and 
the like, the countless multitude of auspicious qualities ; 
the generation at will of all things other than himself, 
whether spiritual or non-spiritual ; various and infinite 
adornment with unsurpassable excellence, singular, uni 
form, and divine. 

Venkata-natha has given the following distribution of 
things : 

" Those who know it have declared the principle to 

be twofold, substance and non-substance ; 
" Substance is dichotomised as unsentient and sentient ; 

the former being the unevolved (avyakta), and 

" The latter is the near (pratyaJc) and the distant 

(pardJc) ; the near being twofold, as either soul 

or the Lord ; 
"The distant is eternal glory and intelligence; the 

other principle some have called the unsentient 

Of these 


" Substance undergoes a plurality of conditions ; the 
origiuant is possessed of goodness and the other 
cords ; 

" Time has the form of years, &c. ; soul is atomic and 
cognisant ; the other spirit is the Lord ; 

"Eternal bliss has been declared as transcending the 
three cords (or modes of phenomenal existence), 
and also as characterised by goodness ; 

" The cognisable manifestation of the cognisant is intel- 

O O 

ligence ; thus are the characteristics of substance 
summarily recounted." 

Of these (soul, not-soul, and the Lord), individual 
spirits, called souls, are different from the Supreme Spirit 
and eternal. Thus the text : Two birds, companions, 
friends, &c. (Kig-Veda, i. 164, 20). Accordingly it is 
stated (in the aphorisms of Kanada, iii. 2, 20), Souls are 
diverse by reason of diversity of conditions. The eternity 
of souls is often spoken of in revelation 

" The soul is neither born, nor dies, nor having been 

shall it again cease to be ; 

" Unborn, unchanging, eternal, this ancient of days is 
not killed when the body is killed " (Bhagavad- 
gita, ii. 20). 

Otherwise (were the soul not eternal) there would follow 
a failure of requital and a fruition (of pleasures and pains) 
unmerited. It has accordingly been said (in the aphorisms 
of Gautama, iii. 25): Because no birth is seen of one who 
is devoid of desire. That the soul is atomic is well known 
from revelation 

" If the hundredth part of a hair be imagined to be 

divided a hundred times, 
" The soul may be supposed a part of that, and yet it is 

capable of infinity." 
And again 

" Soul is of the size of the extremity of the spoke of a 
wheel. Spirit is to be recognised by the intelligence 
as atomic." 


The visible, unsentient world, designated by the term 
not-soul, is divided into three, as the object, the instru 
ment, or the site of fruition. Of this world the efficient 
and substantial cause is the Deity, known under the 
names Purnshottama (best of spirits), Vasudeva (a patrony 
mic of Krishna), and the like. 

" Vasudeva is the supreme absolute spirit, endowed with 
auspicious attributes, 

" The substantial cause, the efficient of the worlds, the 
animator of spirits." 

This same Vasudeva, infinitely compassionate, tender to 
those devoted to him, the Supreme Spirit, with the pur 
pose of bestowing various rewards apportioned to the 
deserts of his votaries in consequence of pastime, exists 
under five modes, distinguished as " adoration " (archd), 
" emanation " (vibhava), " manifestation " (vyuha), " the 
subtile" (sukshma), and the "internal controller." (i.) 
"Adoration" is images, and so forth. (2.) "Emanation" 
is his incarnation, as Bama, and so forth. (3.) His " mani 
festation" is fourfold, as Vasudeva, Sankarshana, Pra- 
dyumna, and Aniruddha. (4.) " The subtile " is the 
entire Supreme Spirit, with six attributes, called Vasu 
deva. His attributes are exemption from sin, and the 
rest. That he is exempt from sin is attested in the Vedic 
text: Passionless, deathless, without sorrow, without 
hunger, desiring truth, true in purpose. (5.) The "in 
ternal controller," the actuator of all spirits, according to 
the text : Who abiding in the soul, rules the soul within. 
When by worshipping each former embodiment a mass of 
sins inimical to the end of the soul (i.e., emancipation) 
have been destroyed, the votary becomes entitled to prac 
tise the worship of each latter embodiment. It has, there 
fore, been said 

" Vasudeva, in his tenderness to his votaries, gives, as 
desired by each, 

" According to the merits of his qualified worshippers, 
large recompense. 


" For that end, in pastime he makes to himself his five 

embodiments ; 
" Images and the like are adoration ; his incarnations 

are emanations ; 

"As Sankarshana, Vasudeva, Pradyumna, Aniruddha, 
his manifestation is to be known to be fourfold ; 
the subtile is the entire six attributes ; 
" That self-same called Vasudeva is styled the Supreme 

Spirit ; 
" The internal controller is declared as residing in the 

soul, the actuator of the soul, 
" Described in a multitude of texts of the Upanishads, 

such as Who abiding in the soul. 
" By the worship of adoration, a man casting off his 

defilement becomes a qualified votary ; 
" By the subsequent worship of emanation, he be 
comes qualified for the worship of manifestation ; 
" By the worship thereafter of the subtile/ he becomes 

able to behold the internal controller. " 
The worship of the Deity is described in the Pancha- 
ratra as consisting of five elements, viz., (i.) the access, (2.) 
the preparation, (3.) oblation, (4.) recitation, (5.) devotion. 
Of these, access is the sweeping, smearing, and so forth, 
of the way to the temple. The preparation is the provision 
of perfumes, flowers, and the like appliances of worship. 
Oblation is worship of the deities. Eecitation is the 
muttered ejaculation of sacred texts, with attention to 
what they mean, the rehearsal of hymns and lauds of 
Vishnu, the commemoration of his names, and study of 
institutes which set forth the truth. Devotion is medita 
tion on the Deity. When the vision of the visible world 
has been brought to a close by knowledge accumulated by 
the merit of such worship, the infinitely compassionate 
Supreme Spirit, tender to his votaries, bestows upon the 
votary devoted to his lord and absorbed in his lord, his 
own sphere infinite and endless, marked by consciousness 


of being like him, from which there is no future return 
(to the sorrows of transmigratory existence). So the 
traditionary text 

"When they have come to me, the high-souled no 
longer undergo future "birth, a receptacle of pain, 
transitory, having attained to the supreme con 

" Vasudeva, having found his votary, bestows upon him 
his own mansion, blissful, undecaying, from whence 
there is no more return." 

After laying up all this in his heart, leaning upon the 
teaching of the great Upanishad, and finding the gloss on 
the Vedanta aphorisms by the venerated Bodhayanacharya 
too prolix, Kamanuja composed a commentary on the 
Sarirakamimansa (or Vedanta theosophy). In this the 
sense of the first aphorism, " Then hence the absolute 
must be desired to be known," is given as follows : The 
word then in this aphorism means, after understanding the 
hitherto-current sacred rites. Thus the glossator writes : 
" After learning the sacred rites," he desires to know the 
absolute. The word hence states the reason, viz., because 
one who has read the Veda and its appendages and under 
stands its meaning is averse from sacred rites, their 
recompense being perishable. The wish to know the 
absolute springs up in one who longs for permanent 
liberation, as being the means of such liberation. By the 
word absolute is designated the Supreme Spirit, from whom 
are essentially excluded all imperfections, who is of illimi 
table excellence, and of innumerable auspicious attributes. 
Since then the knowledge of sacred rites and the perform 
ance of those rites is mediately through engendering dis 
passionateness, and through putting away the defilement 
of the understanding, an instrument of the knowledge of 
the absolute; and knowledge of sacred rites and know 
ledge of the absolute being consequently cause and effect, 
the former and the latter Mimansa constitute one system 
of institutes. On this account the glossator has described 


this system as one with the sixteenfold system of Jaimini. 
That the fruit of sacred rites is perishable, and that of the 
knowledge of the absolute imperishable, has been laid down 
in virtue of Vedic texts, such as : Scanning the spheres 
gained by rites, let him become passionless ; Not wrought 
by the rite performed, accompanied with inference and dis 
junctive reasoning. Eevelation, by censuring each when 
unaccompanied by the other, shows that it is knowledge 
together with works that is efficacious of emancipation, in 
the words : Blind darkness they enter who prefer illusion, 
and a greater darkness still do they enter who delight in 
knowledge only ; knowledge and illusion, he who knows 
these both, he passing beyond death together with illusion, 
tastes immortality by knowledge. Conformably it is said 
in the Paficharatra-rahasya 

"That ocean of compassion, the Lord, tender to his 

"For his worshipper s sake takes five embodiments 

upon him. 
" These are styled Adoration, Emanation, Manifestation, 

the Subtile, the Internal Controller, 
" Eesorting whereto souls attain to successive stages of 

"As a man s sins are worn away by each successive 

" He becomes qualified for the worship of each next 

" Thus day by day, according to religion, revealed and 

" By the aforesaid worship Vasudeva becomes propitious 

to mankind. 
"Hari, when propitiated by devotion in the form of 

" At once brings to a close that illusion which is the 

aggregate of works. 

"Then in souls the essential attributes, from which 
transmigration has vanished, 



"Are manifested, auspicious, omniscience, and the 

" These qualities are common to the emancipated spirits 
and the Lord, 

" Universal efficiency alone among them is peculiar to 
the Deity. 

" Emancipated spirits are ulterior to the infinite absolute, 
which is unsusceptible of aught ulterior ; 

" They enjoy all beatitudes together with that Spirit." 

It is therefore stated that those who suffer the three 
kinds of pain must, for the attainment of immortality, 
investigate the absolute spirit known under such appella 
tions as the Highest Being. According to the maxim : The 
base and the suffix convey the meaning conjointly, and of 
these the meaning of the suffix takes the lead, the notion 
of desire is predominant (in the word jijftdsitavya), and 
desired knowledge is the predicate (in the aphorism, Then 
hence the absolute must be desired to be known). Know 
ledge is cognition designated by such terms as meditation, 
devotion; not the merely superficial knowledge derived 
from verbal communication, such being competent to any 
one who hears a number of words and understands the 
force of each, even without any predication ; in conformity 
with such Vedic texts as : Self indeed it is that is to be 
seen, to be heard, to be thought, to be pondered ; He should 
meditate that it is self alone; Having known, let him 
acquire excellent wisdom; He should know that which 
is beyond knowledge. In these texts " to be heard " is 
explanatory, hearing being understood (but not enounced) 
in the text about sacred study (viz., shadangena vedo dliyeyo 
jneyascha, the Veda, with its six appendages, is to be 
studied and known) ; so that a man who has studied the 
Veda must of his own accord, in acquiring the Veda and 
its appendages, engage in " hearing," in order to ascertain 
the sense by examining it and the occasion of its enounce- 
ment. The term " to be thought " (or " to be inferred ") 
is also explanatory, cogitation (or inference) being under- 


stood as the complementary meaning of hearing, according 
to the aphorism: Before its signification is attained the 
system is significant. Meditation is a reminiscence con 
sisting of an unbroken succession of reminiscences like a 
stream of oil, it being revealed in the text, in continuity 
of reminiscence there is a solution of all knots, that 
it is unintermittent reminiscence that is the means of 
emancipation. And this reminiscence is tantamount to 

" Cut is his heart s knot, solved are all his doubts, 

" And exhausted are all his works, when he has seen 

the Highest and Lowest," 

because he becomes one with that Supreme. So also in 
the words, Self indeed is to be seen, it is predicated of this 
reminiscence that it is an intuition. Reminiscence be 
comes intuitional through the vivacity of the representa 
tions. The author of the Vakya has treated of all this in 
detail in the passage beginning Cognition is meditation. 
The characters of this meditation are laid out in the text : 
This soul is not attainable by exposition, nor by wisdom, 
nor by much learning ; Whom God chooses by him God 
may be attained. To him this self unfolds its own 
nature. For it is that which is dearest which is choice- 
worthy, and as the soul finds itself most dear, so the Lord 
is of Himself most dear, as was declared by the Lord 

" To them always devoted, who worship me with love, 
" I give the devotion of understanding whereby they 

come to me." 
And again 

"That Supreme Spirit, Arjuna, is attainable by faith 


But devotion (or faith) is a kind of cognition which 
admits no other motive than the illimitable beatitude, and 
is free from all other desires ; and the attainment of this 
devotion is by discrimination and other means. As is 
said by the author of the Vakya: Attainment thereof 


results from discrimination (viveJca), exemption (vimokd), 
practice (alJiydsa) , observance (kriyd), excellence (Jcalydna)> 
freedom from despondency (anavasdda), satisfaction (anud- 
dharsha), according to the equivalence (of the definition), 
and the explication (of these terms). Of these means, 
discrimination is purity of nature, resultant from eating 
undefiled food, and the explication (of discrimination) is 
From purity of diet, purity of understanding, and by 
purity of understanding the unintermittent reminiscence. 
Exemption is non-attachment to sensuous desires ; the 
explication being, Let the quietist meditate. Practice is 
reiteration ; and of this a traditionary explication is quoted 
(from the Bhagavad-gita) by (Eamanuja) the author of 
the commentary : For ever modified by the modes thereof. 
Observance is the performance of rites enjoined in revela 
tion and tradition according to one s ability ; the explica 
tion being (the Vedic text), He who has performed rites 
is the best of those that know the supreme. The excel 
lences are veracity, integrity, clemency, charity (alms 
giving), and the like ; the explication being, It is attained 
by veracity. Freedom from despondency is the contrary 
of dejection ; the explication being, This soul is not attained 
by the faint-hearted. Satisfaction is the contentment 
which arises from the contrary of dejection ; the explica 
tion being, Quiescent, self-subdued. It has thus been 
shown that by the devotion of one in whom the darkness 
has been dispelled by the grace of the Supreme Spirit, 
propitiated by certain rites and observances, which devo 
tion is meditation transformed into a presentative mani 
festation of soul, without ulterior motive, as incessantly 
and inimitably desired, the sphere of the Supreme Spirit 
(Vaikuntha) is attained. Thus Yamuna says : Attainable 
by the final and absolute devotion of faith in one internally 
purified by both (works and knowledge) ; that is, in one 
whose internal organ is rectified by the devotion of works 
and knowledge. 

In anticipation of the inquiry, But what absolute is to 


be desired to be known ? the definition is given (in the 
second aphorism). From which the genesis, and so forth, 
of this. The genesis, and so forth, the creation (emana 
tion), sustentation, and retractation (of the universe). 
The purport of the aphorism is that the emanation, sus 
tentation, and retractation of this universe, inconceivably 
multiform in its structure, and interspersed with souls, 
from Brahma to a tuft of grass, of determinate place, 
time, and fruition, is from this same universal Lord, whose 
essence is contrary to all qualities which should be escaped 
from, of illimitable excellences, such as indefeasible voli 
tion, and of innumerable auspicious attributes, omniscient, 
and omnipotent. 

In anticipation of the further inquiry, What proof is 
there of an absolute of this nature ? It is stated that the 
system of institutes itself is the evidence (in the third 
aphorism) : Because it has its source from the system. 
To have its source from the system is to be that whereof 
the cause or evidence is the system. The system, then, is 
the source (or evidence) of the absolute, as being the cause 
of knowing the self, which is the cause of knowing the 
absolute. Nor is the suspicion possible that the absolute 
may be reached by some other form of evidence. For 
perception can have no conversancy about the absolute 
since it is supersensible. Nor can inference, for the 
illation, the ocean, and the rest, must have a maker, be 
cause it is an effect like a water-pot, is worth about as 
much as a rotten pumpkin. It is evinced that it is such 
texts as, Whence also these elements, that prove the 
existence of the absolute thus described. 

Though the absolute (it may be objected) be unsuscep 
tible of any other kind of proof, the system, did it not 
refer to activity and cessation of activity, could not posit 
the absolute aforesaid. To avoid by anticipation any 
queries on this point, it is stated (in the fourth aphorism) : 
But that is from the construction. This is intended to 
exclude the doubt anticipated. The evidence, then, of the 


system is the only evidence that can be given of the 
absolute. Why? Because of the construction, that is 
because the absolute, that is, the highest end for man, is 
construed as the subject (of the first aphorism, viz., Then 
thence the absolute is to be desired to be known). More 
over, a sentence which has nothing to do either with acti 
vity or with cessation of activity is not therefore void of 
purpose, for we observe that sentences merely declaratory 
of the nature of things, such as, A son is born to you, This 
is not a snake, convey a purpose, viz., the cessation of joy 
or of fear. Thus there is nothing unaccounted for. We 
have here given only a general indication. The details 
may be learnt from the original (viz., Eamanuja s Bhashya 
on the Vedanta aphorisms) ; we therefore decline a further 
treatment, apprehensive of prolixity; and thus all is 
clear. 1 A. E. G. 

1 For further details respecting tva-muktdvaU was printed in the 

Ramdnuja and his system, see Wil- Pandit for September 1871; but the 

son s Works, vol. i. pp. 34-46 ; and lines quoted in p. 73 are not found 

Banerjea s Dialogues, ix. The Tat- there. 



ANANDA-TIRTHA (Piirna-prajna, or Madhva) rejected this 
same Ramanuja system, because, though like his own 
views, it teaches the atomic size of the soul, the servitude 
of the soul, the existence of the Veda without any per 
sonal author, the authenticity of the Veda, the self- evidence 
of the instruments of knowledge, the triad of evidences, 
dependency upon the Pancha-ratra, the reality of plurality 
in the universe, and so forth, yet, in accepting three 
hypotheses as to reciprocally contradictory divisions, &c., 
it coincides with the tenets of the Jainas. Showing that 
He is soul, That art thou, and a number of other texts of 
the Upanishads bear a different import under a different 
explanation, he set up a new system under the guise of a 
new explication of the Brahma-Mimansa (or Vedanta). 

For in his doctrine ultimate principles are dichotomised 
into independent and dependent; as it is stated in the 
Tattva-viveka : 

"Independent and dependent, two principles are re 
ceived ; 

" The independent is Vishnu the Lord, exempt from 
imperfections, and of inexhaustible excellences." 

Here it will be urged (by the Advaita-vadins) : Why 
predicate of the absolute these inexhaustible excellences 
in the teeth of the Upanishads, which lay down that the 
absolute principle is void of homogeneity and hetero 
geneity, and of all plurality in itself? To this be it 


replied: Not so, for these texts of the Upanishads, as 
contradictory of many proofs positive of duality, cannot 
afford proof of universal unity ; perception, for example, 
in the consciousness, This is different from that, pronounces 
a difference between things, blue and yellow, and so forth. 
The opponent will rejoin : Do you hold that perception is 
cognisant of a perceptional difference, or of a difference 
constituted by the thing and its opposite? The former 
alternative will not hold : for without a cognition of the 
thing and its opposite, the recognition of the difference, 
which presupposes such a cognition, will be impossible. 
On the latter alternative it must be asked, Is the appre 
hension of the difference preceded by an apprehension of 
the thing and its contrary, or are all the three (the thing, 
its contrary, and the contrariety) simultaneously appre 
hended ? It cannot be thus preceded, for the operation 
of the intellect is without delay (or without successive 
steps), and there would also result a logical seesaw (appre 
hension of the difference presupposing apprehension of 
the thing and its contrary, and apprehension of the thing 
and its contrary presupposing apprehension of the differ 
ence). Nor can there be a simultaneous apprehension (of 
the thing, its contrary, and the difference) ; for cognitions 
related as cause and effect cannot be simultaneous, and 
the cognition of the thing is the cause of the recognition 
of the difference; the causal relation between the two 
being recognised by a concomitance and non-concomitance 
(mutual exclusion), the difference not being cognised even 
when the thing is present, without a cognition of its absent 
contrary. The perception of difference, therefore (the 
opponent concludes), is not easily admissible. To this let 
the reply be as follows : Are these objections proclaimed 
against one who maintains a difference identical with the 
things themselves, or against one who maintains a differ 
ence between things as the subjects of attributes ? In the 
former case, you will be, as the saying runs, punishing a 
respectable Brahman for the offence of a thief, the objec- 


tions you adduce being irrelevant. If it be urged that if 
it is the essence of the thing that is the difference, then 
it will no longer require a contrary counterpart; but if 
difference presuppose a contrary counterpart, it will exist 
everywhere ; this statement must be disallowed, for while 
the essence of a thing is first known as different from 
everything else, the determinate usage (name and notion) 
may be shown to depend upon a contrary counterpart; 
for example, the essence of a thing so far as constituted 
by its dimensions is first cognised, and afterwards it be 
comes the object of some determinate judgment, as long or 
short in relation to some particular counterpart (or con 
trasted object). Accordingly, it is said in the Vishnu- 
tattva-nirnaya : " Difference is not proved to exist by the 
relation of determinant and determinate ; for this relation 
of determinant and determinate (or predicate and subject) 
presupposes difference; and if difference were proved to 
depend upon the thing and its counterpart, and the thing 
and its counterpart to presuppose difference, difference as 
involving a logical circle could not be accounted for ; but 
difference is itself a real predicament (or ultimate entity). 
Tor this reason (viz., because difference is a thing) it is 
that men in quest of a cow do not act (as if they had 
found her) when they see a gayal, and do not recall the 
word cow. Nor let it be objected that (if difference be a 
real entity and as such perceived) on seeing a mixture of 
milk and water, there would be a presentation of differ 
ence ; for the absence of any manifestation of, and judg 
ment about, the difference, may be accounted for by the 
force of (the same) obstructives (as hinder the perception 
of other things), viz., aggregation of similars and the rest. 
Thus it has been said (in the Sankhya-karika, v. vii.) 

" From too great remoteness, from too great nearness, 
from defect in the organs, from instability of the 
common sensory, 

"From subtilty, from interposition, from being over 
powered, and from aggregation of similars." 


There is no perception respectively of a tree and the 
like on the peak of a mountain, because of its too great 
remoteness ; of collyrium applied to the eyes, and so forth, 
because of too great proximity ; of lightning and the like, 
because of a defect in the organs; of a jar or the like 
in broad daylight, by one whose common sensory is be 
wildered by lust and other passions, because of instability 
of the common sensory ; of an atom and the like, because 
of their subtility ; of things behind a wall, and so forth, 
because of interposition ; of the light of a lamp and the 
like, in the day-time, because of its being overpowered ; 
of milk and water, because of the aggregation of similars. 

Or let the hypothesis of difference in qualities be 
granted, and no harm is done ; for given the apprehension 
of a subject of attributes and of its contrary, the presenta 
tion of difference in their modes is possible. Nor let it be 
supposed that on the hypothesis of difference in the modes 
of things, as each difference must be different from some 
ulterior difference, there will result an embarrassing pro 
gression to infinity, there being no occasion for the 
occurrence of the said ulterior difference, inasmuch as we 
do not observe that men think and say that two things are 
different as differenced from the different. Nor can an 
ulterior difference be inferred from the first difference, for 
there being no difference to serve as the example in such 
inference, there cannot but be a non-occurrence of infer 
ence. And thus it must be allowed that in raising the 
objection you have begged for a little oil-cake, and have 
had to give us gallons of oil. If there be no difference for 
the example the inference cannot emerge. The bride is 
not married for the destruction of the bridegroom. There 
being, then, no fundamental difficulty, this infinite pro 
gression presents no trouble. 

Difference (duality) is also ascertained by inference. 
Thus the Supreme Lord differs from the individual soul 
as the object of its obedience ; and he who is to be obeyed 
by any person differs from that person, a king, for in- 


stance, from his attendant. For men, desiring as they do 
the end of man, Let me have pleasure, let me not have 
the slightest pain, if they covet the position of their lord, 
do not become objects of his favour, nay, rather, they be 
come recipients of all kinds of evil. He who asserts his 
own inferiority and the excellence of his superior, he it 
is who is to be commended; and the gratified superior 
grants his eulogist his desire. Therefore it has been 
said : 

"Kings destroy those who assert themselves to be 

"And grant to those who proclaim their kingly pre 
eminence all that they desire." 

Thus the statement of those (Advaita-vadins) in their 
thirst to be one with the Supreme Lord, that the supreme 
excellence of Vishnu is like a mirage, is as if they were to 
cut off their tongues in trying to get a fine plantain, since 
it results that through offending this supreme Vishnu they 
must enter into the hell of blind darkness (andha-tamasa). 
The same thing is laid down by Madhya-mandira in the 
Mahabharata-tatparya-nirnaya : 

" Daityas, enemies of the eternal, Vishnu s anger is 
waxed great ; 

" He hurls the Daityas into the blind darkness, because 
they decide blindly." 

This service (or obedience of which we have spoken) is 
trichotomised into (i.) stigmatisation, (2.) imposition of 
names, (3.) worship. 

Of these, (i.) stigmatisation is (the branding upon one 
self) of. the weapons of Narayana (or Vishnu) as a memorial 
of him, and as a means of attaining the end which is 
needful (emancipation). Thus the sequel of the Sakalya- 
samhita : 

"The man who bears branded in him the discus of 
the immortal Vishnu, which is the might of the 

" He, shaking off his guilt, goes to the heaven (Vaikun- 


tha) which ascetics, whose desires are passed away, 
enter into : 
" The discus Sudarsana by which, uplifted in his arm, 

the gods entered that heaven ; 

" Marked wherewith the Manus projected the emana- 
tion of the world, that weapon Brahmans wear 
(stamped upon them) ; 
" Stigmatised wherewith they go to the supreme sphere 

of Vishnu ; 
" Marked with the stigmas of the wide-striding (Vishnu), 

let us become beatified." 

Again, the Taittiriyaka Upanishad says : " He whose 
body is not branded, is raw, and tastes it not : votaries 
bearing it attain thereto." The particular parts to be 
branded are specified in the Agneya-purana : 

" On his right hand let the Brahman wear Sudarsana, 
"On his left the conch-shell: thus have those who 

know the Veda declared." 

In another passage is given the invocation to be recited 
on being branded with the discus : 

" Sudarsana, brightly blazing, effulgent as ten million 

" Show unto me, blind with ignorance, the everlasting 

way of Vishnu. 
" Thou aforetime sprangest from the sea, brandished in 

the hand of Vishnu, 
"Adored by all the gods; Panchajanya, to thee be 


(2.) Imposition of names is the appellation of sons and 
others by such names as Kesava, as a continual memorial 
of the name of the Supreme Lord. 

(3.) Worship is of ten kinds, viz., with the voice, (i.) 
veracity, (2.) usefulness, (3.) kindliness, (4.) sacred study ; 
with the body, (5.) almsgiving, (6.) defence, (7.) protection ; 
with the common sensory, (8.) mercy, (9.) longing, and 
(10.) faith. Worship is the dedication to Narayana of 
each of these as it is realised. Thus it has been said : 


" Stigmatisation, imposition of names, worship; the last 
is of ten kinds." 

Difference (or duality between the Supreme Being and 
the universe) may also be inferred from cognisability and 
other marks. So also difference (or duality) may be 
understood from revelation, from texts setting out duality 
in emancipation and beatitude, such as : " All rejoice over 
truth attained; truthful, and celebrating the gift of the 
divine Indra, they recount his glory ; " " Sarva, among those 
that know the truth, Brahman, is in the universe, true 
spirit ; true is individual spirit ; truth is duality, truth 
is duality, in me is illusion, in me illusion, in me 

Again : 

"After attaining this knowledge, becoming like unto 

"In creation they are not born again, in retractation 
they perish not" (Bhagavad-gita, xiv. 2). 

According also to such aphorisms as, " Excepting cos- 
mical operation because of occasion, and because of non- 

Nor should suggestion be made that individual spirit 
is God in virtue of the text, He that knows the absolute 
becomes the absolute; for this text is hyperbolically 
eulogistic, like the text, Worshipping a Brahman devoutly 
a Sudra becomes a Brahman, i.e., becomes exalted. 

If any one urge that according to the text : 

"If the universe existed it would doubtless come to an 


this duality is merely illusory, and in reality a unity, 
and that duality is learnt to be illusorily imagined ; it may 
be replied : What you say is true, but you do not under 
stand its meaning ; for the real meaning is, If this world 
had been produced, it would, without doubt, come to an 
end; therefore this universe is from everlasting, a five 
fold dual universe ; and it is not non-existent, because 
it is mere illusion. Illusion is denned to be the will of 


the Lord, in virtue of the testimony of many such pas 
sages as : 

" The great illusion, ignorance, necessity, the bewilder 
" The origiriant, ideation, thus is thy will called, 

" The originant, because it originates greatly ; ideation, 

because it produces ideas ; 
The illusion of Hari, who is called a, is termed (a-vidyd) 

ignorance : 
" Styled (mdyd) illusion, because it is pre-eminent, for 

the name mdyd is used of the pre-eminent ; 
" The excellent knowledge of Vishnu is called, though 

one only, by these names ; 

" For Hari is excellent knowledge, and this is character 
ised by spontaneous beatitude." 

That in which this excellent knowledge produces know 
ledge and effects sustentation thereof, that is pure illusion, 
as known and sustained, therefore by the Supreme Lord 
duality is not illusorily imagined. For in the Lord illu 
sory imagination of the universe is not possible, illusory 
imagination arising from non-perception of differences 
(which as an imperfection is inconsistent with the divine 

If it be asked how then that (illusory duality) is pre 
dicated, the answer is that in reality there is a non-duality, 
that is in reality, Vishnu being better than all else, has 
no equal and no superior. Accordingly, the grand revela 
tion : 

" A difference between soul and the Lord, a difference 

between the unsentient and the Lord, 
" A difference among souls, and a difference of the 

unsentient and the soul each from the other. 
" Also the difference of unsentient things from one 

another, the world with its five divisions. 
" This same is real and from all eternity ; if it had had 
a beginning it would have an end : 


" Whereas it does not come to an end ; and it is not 

illusorily imagined : 
" For if it were imagined it would cease, but it never 

" That there is no duality is therefore the doctrine of 

those that lack knowledge ; 
" For this the doctrine of those that have knowledge is 

known and sustained by Vishnu." 
The purpose, then, of all revelations is to set out the 
supreme excellence of Vishnu. With this in view the 
Lord declared : 

" Two are these persons in the universe, the perishable 

and the imperishable ; 
" The perishable is all the elements, the imperishable is 

the unmodified. 
"The other, the most excellent person, called the 

Supreme Spirit, . 
" Is the undecaying Lord, who pervading sustains the 

three worlds. 
" Since transcending the perishable, I am more excellent 

than the imperishable (soul), 
" Hence I am celebrated among men and in the Veda 

as the best of persons (Purushottama) ; 
"He who uninfatuated knows me thus the best of 

persons, he all-knowing worships me in every wise. 
" Thus this most mysterious institute is declared, blame 
less (Arjuna) : 
" Knowing this a man may be wise, and may have done 

what he has to do, Bharata" (Bhagavad-gita, 

xv. 16-20). 

So in the Maha-varaha 
11 The primary purport of all the Vedas relates to the 

supreme spouse of Sri ; 
" Its purport regarding the excellence of any other deity 

must be subordinate." 

It is reasonable that the primary purport should regard 
the supreme excellence of Vishnu. For emancipation is 


the highest end of all men, according to the text of the 
Bhallaveya Upanishad : While merit, wealth, and enjoy 
ment are transitory, emancipation is eternal ; therefore a 
wise man should strive unceasingly to attain thereto. 
And emancipation is not won without the grace of Vishnu, 
according to the text of the Narayana Upanishad : Through 
whose grace is the highest state, through whose essence he 
is liberated from transmigration, while inferior men pro 
pitiating the divinities are not emancipated ; the supreme 
object of discernment to those who desire to be liberated 
from this snare of works. According also to the words of 
the Vishnu-purana 

" If he be propitiated, what may not here be won ? 
Enough of all wealth and enjoyments. These are scanty 
enough. On climbing the tree of the supreme essence, 
without doubt a man attains to the fruit of emancipa 

And it is declared that the grace of Vishnu is won only 
through the knowledge of his excellence, not through the 
knowledge of non-duality. Nor is there in this doctrine 
any connection with texts declaratory of the identity (of 
personal and impersonal spirit) such as, That art thou (for 
this pretended identity) is mere babbling from ignorance 
of the real purport. 

"The word That, when undetermined, designates the 
eternally unknown, 

" The word Thou designates a knowable entity; how can 
these be one ? " 

And this text (That art thou) indicates similarity (not 
identity) like the text, The sun is the sacrificial post. 
Thus the grand revelation : 

"The ultimate unity of the individual soul is either 
similarity of cognition, 

" Or entrance into the same place, or in relation to the 
place of the individual ; 

" Not essential unity, for even when it is emancipated 
it is different, 


" The difference being independence and completeness 
(in the Supreme Spirit), and smallness and depend 
ence (in the individual spirit)." 

Or to propose another explanation of the text, Atmd 
tat tvam asi, That art thou, it may be divided, dtmd 
atat tvam asi. He alone is soul as possessing indepen 
dence and other attributes, and thou art not-that (atat) 
as wanting those attributes ; and thus the doctrine of 
unity is utterly expelled. Thus it has been said : 
" Or the division may be Atat tvam, and thus unity will 

be well got rid of." 

According, therefore, to the Tattva-vada-rahasya, the 
words in the nine examples (in the Chhandogya Upani- 
shad), He like a bird tied with a .string, &c., teach unity 
with the view of giving an example of non-duality. 
Accordingly the Mahopanishad : 

" Like a bird and the string ; like the juices of various 

trees ; 

" Like rivers and the sea ; like fresh and salt water ; 
" Like a robber and the robbed ; like a man and his 

energy ; 

" So are soul and the Lord diverse, for ever different. 
" Nevertheless from subtilty (or imperceptibility) of 

form, the supreme Hari 
"Is not seen by the dim-sighted to be other than the 

individual spirit, though he is its actuator; 
" On knowing their diversity a man is emancipated : 

otherwise he is bound." 
And again 

" Brahma, Siva, and the greatest of the gods decay with 

the decay of their bodies ; 
"Greater than these is Hari, undecaying, because his 

body is for the sustentation of Lakshmi. 
" By reason of all his attributes, independence, power, 

knowledge, pleasure, and the rest, 
" All they, all the deities, are in unlimited obedience to 



And again : 

Knowing Vishnu, full of all excellences, the soul, 
exempted from transmigration, 

" Eejoices in his presence for ever, enjoying painless 

Vishnu is the refuge of liberated souls, and their 
supreme ruler. 

" Obedient to him are they for ever ; he is the Lord." 

That by knowledge of one thing there is knowledge of 
all things may be evinced from its supremacy and causality, 
not from the falsity of all things. For knowledge of the 
false cannot be brought about by knowledge of real exist 
ence. As we see the current assurance and expression 
that by knowing or not knowing its chief men a village 
is known or not known ; and as when the father the cause 
is known, a man knows the son; (so by knowing the 
supreme and the cause, the inferior and the effect is known). 
Otherwise (on the doctrine of the Advaita-vadins that the 
world is false and illusory) the words one and lump in the 
text, By one lump of clay, fair sir, all that is made of clay 
is recognised, would be used to no purpose, for the text 
must be completed by supplying the words, By reason of 
clay recognised. For the text, Utterance with the voice, 
modification, name, clay (or other determinate object), 
these alone are real, cannot be assumed to impart the 
falsity of things made ; the reality of these being admitted, 
for what is meant is, that of which utterance with the 
voice is a modification, is unmodified, eternal ; and a name 
such as clay, such speech is true. Otherwise it would 
result that the words name and alone would be otiose. 
There is no proof anywhere, then, that the world is unreal. 
Besides (we would ask) is the statement that the world is 
false itself true or false. If the statement is true, there 
is a violation of a real non-duality. If the statement is 
untrue, it follows that the world is true. 

Perhaps it may be objected that this dilemma is a kind 
of fallacious reasoning, like the dilemma : Is transitoriness 


permanent or transitory ? There is a difficulty in either 
case. As it is said by the author of the Nyaya-nirvana : 
The proof of the permanence of the transitory, as being 
both permanent and transitory, is a paralogism. And in 
the Tarkika-raksha 

" When a mode cannot be evinced to be either such and 
such, or not such and such, 

" The denial of a subject characterised by such a mode 
is called Nitya-sama. 

With the implied mention of this same technical ex 
pression it is stated in the Prabodha-siddhi : Equality of 
characteristic modes results from significancy. If it be 
said, This then is a valid rejoinder, we reply, This is a 
mere scaring of the uninstructed, for the source of fallacy 
has not been pointed out. This is twofold, general and 
particular : of these, the former is self-destructive, and the 
latter is of three kinds, defect of a requisite element, 
excess of an element not requisite, and residence in that 
which is not the subjicible subject. Of these (two forms 
of the fallacy), the general form is not suspected, no self- 
pervasion being observed in the dilemma in question (viz., 
Is the statement that the world is unreal itself true or 
false? &c.) So likewise the particular; for if a water-jar 
be said to be non-existent, the affirmation of its non- 
existence is equally applicable to the water-jar as that of 
its existence. 

If you reply: We accept the unreality (or falsity) of 
the world, not its non-existence; this reply is about as 
wise as the procedure of the carter who will lose his head 
rather than pay a hundred pieces of money, but will at 
once give five score; for falsity and non-existence are 
synonymous. We dismiss further prolixity. 

The meaning of the first aphorism, viz., Then hence the 
absolute is to be desired to be known, is as follows : The 
word then is allowed to purport auspiciousness, and to 
designate subsequency to the qualification (of the aspirant). 
The word hence indicates a reason. 


Accordingly it is stated in the Garuda-purana : 
"All the aphorisms begin with the words Then and 
Hence regularly ; what then is the reason of this ? 
" And what is the sense of those words, sage ? Why- 
are those the most excellent ? 
" Tell me this, Brahma, that I may know it truly." 
Thus addressed by Narada, the most excellent Brahma 
replied : 

" The word Then is used of subsequency and of com 
petency, and in an auspicious sense, 
"And the word Thence is employed to indicate the 


It is laid down that we must institute inquiries about 
the absolute, because emancipation is not attained with 
out the grace of Narayana, and his grace is not attained 
without knowledge. The absolute, about which the in 
quiry is to be instituted, is described in the words (of the 
second aphorism) : From which the genesis, and so forth, 
of this. The meaning of the sentence is that the absolute 
is that from which result emanation, sustentation, and 
retractation ; according to the words of the Skanda- 

" He is Hari the sole ruler, the spirit from whom are 
emanation, sustentation, retractation, necessity, 
knowledge, involution (in illusion), and bondage 
and liberation ; 

and according to such Vedic texts, From which are these. 
The evidence adducible for this is described (in the third 
aphorism) : Because it has its source from the system. 
That the absolute should be reached by way of inference 
is rejected by such texts as, He that knows not the Veda 
cogitates not that mighty one; Him described in the 
Upanishacls. Inference, moreover, is not by itself autho 
ritative, as is said in the Kaurma-purana 

" Inference, unaccompanied by revelation, in no case 
" Can definitely prove a matter, nor can any other form 
of evidence ; 


"Whatsoever other form of evidence, companioned by 

revelation and tradition, 
"Acquires the rank of probation, about this there can 

be no hesitation." 

What a Sastra (or system of sacred institutes) is, has 
been stated in the Skanda-purana : 

"The Pig-veda, the Yajur-veda, the Sama-veda, the 
Atharva-veda, the Mahabharata, the Pancha-ratra, and 
the original Kamayana, are called Sastras. 

"That also which is conformable to these is called 

"Any aggregate of composition other than this is a 


According, then, to the rule that the sense of the sacred 
institutes is not to be taken from other sources than these, 
the Monist view, viz., that the purport of the texts of the 
Veda relates not to the duality learnt from those but to 
non-duality, is rejected: for as there is no proof of a God 
from inference, so there is no proof of the duality between 
God and other things from inference. Therefore there 
can be in these texts no mere explanation of such duality, 
and the texts must be understood to indicate the duality. 
Hence it is that it has said : 

" I ever laud Narayana, the one being to be known from 
genuine revelation, who transcends the perishable 
and the imperishable, without imperfections, and 
of inexhaustible excellences." 

It has thus been evinced that the sacred institutes are 
the evidence of (the existence of) this (ultimate reality, 
Brahman). (The fourth aphorism is).: But that is from 
the construction. In regard to this, the commencement 
and other elements are stated to be the marks of the con 
struction, in the Brihat-samhita : 

" Commencement, conclusion, reiteration, novelty, profit, 
eulogy, and demonstration, are the marks by which 
the purport is ascertained." 
It is thus stated that in accordance with the purport of 


the Upanishads the absolute is to be apprehended only 
from the sacred institutes. We have here given merely 
a general indication. What remains may be sought from 
the Anandatirtha-bhashya-vyakhyana (or exposition of 
the Commentary of Ananda-tirtha). We desist for fear 
of giving an undue prolixity to our treatise. This mystery 
was promulgated by Purna-prajna Madhya-mandira, who 
esteemed himself the third incarnation of Vayu : 
" The first was Hanumat, the second Bhima, 
" The third Purna-prajiia, the worker of the work of the 


After expressing the same idea in various passages, he 
has written the following stanza at the conclusion of his 
work : 

" That whereof the three divine forms are declared in 

the text of the Veda, sufficiently 
" Has that been set forth ; this is the whole majesty in 

the splendour of the Veda ; 

"The first incarnation of the Wind-god was he that 
bowed to the words of Kama (Hanumat) ; the 
second was Bhima ; 
" By this Madhva, who is the third, this book has been 

composed in regard to Kesava." 

The import of this stanza may be learnt by considering 
various Vedic texts. 

The purport of this is that Vishnu is the principle 
above all others in every system of sacred institutes. 
Thus all is clear. 1 A. E. G. 

i For a further account of Ananda- tary on the Brahma-sutras has been 
tirtha or Madhva see Wilson, Works, printed in Calcutta. 
vol. i. pp. 138-150. His Commen- 

( 103 ) 



CERTAIN Mahesvaras disapprove of this doctrine of the 
Vaishnavas known by its technicalities of the servitude of 
souls and the like, inasmuch as bringing with it the pains 
of dependence upon another, it cannot be a means of 
cessation of pain and other desired ends. They recognise 
as stringent such arguments as, Those depending on another 
and longing for independence do not become emancipated, 
because they still depend upon another, being destitute of 
independence like ourselves and others; and, Liberated 
spirits possess the attributes of the Supreme Deity, because 
at the same time, that they are spirits they are free from 
the germ of every pain as the Supreme Deity is. Recog 
nising these arguments, these Mahesvaras adopt the Pdsu- 
pata system, which is conversant about the exposition of 
five categories, as the means to the highest end of man. 
In this system the first aphorism is : Now then we shall 
expound the Pasupata union and rites of Pasupati. The 
meaning is as follows : The word now refers to some 
thing antecedent, and this something antecedent is the 
disciple s interrogation of the spiritual teacher. The 
nature of a spiritual teacher is explicated in the Gana- 
karika : 

" But there are eight pentads to be known, and a group, 
one with three factors ; 

" He that knows this ninefold aggregate is a self-puri 
fier, a spiritual guide. 


"The acquisitions, the impurities, the expedients, the 
localities, the perseverance, the purifications, 

" The initiations, and the powers, are the eight pentads ; 
and there are three functions." 

The employment in the above line of the neuter numeral 
three (trtni), instead of the feminine three (tisrah), is a 
Vedic construction. 

(a.) Acquisition is the fruit of an expedient while realis 
ing, and is divided into five members, viz., knowledge, 
penance, permanence of the body, constancy, and purity. 
Thus Haradattacharya says : Knowledge, penance, per 
manence, constancy, and purity as the fifth. 

(&.) Impurity is an evil condition pertaining to the soul. 
This is of five kinds, false conception and the rest. Thus 
Haradatta also says : 

" False conception, demerit, attachment, interestedness, 
and falling, 

"These five, the root of bondage, are in this system 
especially to be shunned/ 

(c.) An expedient is a means of purifying the aspirant 
to liberation. 

These expedients are of five kinds, use of habitation, and 
the rest. Thus he also says : 

" Use of habitation, pious muttering, meditation, con 
stant recollection of Eudra, 

"And apprehension, are determined to be the five ex 
pedients of acquirements." 

(d.) Locality is that by which, after studying the cate 
gories, the aspirant attains increase of knowledge and 
austerity, viz., spiritual teachers and the rest. Thus he 
says : 

" The spiritual teachers, a cavern, a special place, the 
burning-ground, and Eudra only." 

(e.) Perseverance is the endurance in one or other of 
these pentads until the attainment of the desired end, and 
is distributed into the differenced and the rest. Thus it is 
said : 


"The differenced, the undifferenced, muttering, accep 
tance, and devotion as the fifth." 

(/.) Purification is the putting away, once for all, of 
false conception and the other four impurities. It is dis 
tributed into five species according to the five things to be 
put away. Thus it is said 

"The loss of ignorance, of demerit, of attachment, of 

" And of falling, is declared to be the fivefold purifica 
tion of the state of bondage." 

(#.) The five initiations are thus enumerated : 

" The material, the proper time, the rite, the image, and 
the spiritual guide as the fifth." 

(A.) The five powers are as follow : 

" Devotion to the spiritual guide, clearness of intellect, 
conquest of pleasure and pain, 

" Merit and carefulness, are declared the five heads of 

The three functions are the modes of earning daily food 
consistent with propriety, for the diminution of the five 
impurities, viz., mendicancy, living upon alms, and living 
upon what chance supplies. All the rest is to be found 
in the standard words of this sect. 

In the first aphorism above recited, the word now 
serves to introduce the exposition of the termination of 
pain (or emancipation), that being the object of the 
interrogation about the putting away of pain personal, 
physical, and hyperphysical. By the word pam we are 
to understand the effect (or created world), the word desig 
nating that which is dependent on something ulterior. 
By the word pati we are to understand the cause (or 
prineipium), the word designating the Lord, who is the 
cause of the universe, the pati, or ruler. The meaning of 
the words sacrifices and rites every one knows. 

In this system the cessation of pain is of two kinds, 
impersonal and personal. Of these, the impersonal con 
sists in the absolute extirpation of all pains ; the personal 


in supremacy consisting of the visual and active powers. 
Of these two powers the visual, while only one power, is, 
according to its diversity of objects, indirectly describable 
as of five kinds, vision, audition, cogitation, discrimination, 
and omniscience. Of these five, vision is cognition of 
every kind of visual, tactual, and other sensible objects, 
though imperceptible, intercepted, or remote. Audition 
is cognition of principles, conversant about all articulate 
sounds. Cogitation is cognition of principles, conversant 
about all kinds of thoughts. Discrimination is cognition of 
principles conversant about the whole system of institutes, 
according to the text and according to its significance. 
Omniscience is cognition of principles ever arising and 
pervaded by truth, relative to all matters declared or not 
declared, summary or in detail, classified and specialised. 
Such is this intellectual power. 

The active power, though one only, is indirectly describ 
able as of three kinds, the possession of the swiftness of 
thought, the power of assuming forms at will, and the 
faculty of expatiation. Of these, the possession of the 
swiftness of thought is ability to act with unsurpassable 
celerity. The power of assuming forms at will is the 
faculty of employing at pleasure, and irrespective of 
the efficacy of works, the organs similar and dissimilar 
of an infinity of organisms. The faculty of expatiation 
is the possession of transcendent supremacy even when 
such organs are not employed. Such is this active 

All that is effected or educed, depending on something 
ulterior, it is threefold, sentiency, the insentient, and the 
sentient. Of these, sentiency is the attribute of the sen- 
tients. It is of two degrees according to its nature as 
cognitive or incognitive. Cognitive sentiency is dichoto 
mised as proceeding discriminately and as proceeding 
indiscriminately. The discriminate procedure, manifest 
able by the instruments of knowledge, is called the cogita 
tive. Tor by the cogitant organ every sentient being is 


cognisant of objects in general, discriminated or not dis 
criminated, when irradiated by the light which is identical 
with the external things. The incognitive sentiency, again, 
is either characterised or not characterised by the objects 
of the sentient soul. 

The insentient, which while unconscious is dependent 
on the conscious, is of two kinds, as styled the effect and 
as styled the cause. The insentient, styled the effect, is 
of ten kinds, viz., the earth and the other four elements, 
and their qualities, colour, and the rest. The insentient, 
called the causal insentient, is of thirteen kinds, viz., the 
five organs of cognition, the five organs of action, and the 
three internal organs, intellect, the egoising principle, and 
the cogitant principle, which have for their respective 
functions ascertainment, the illusive identification of self 
with not-self, and determination. 

The sentient spirit, that to which transmigratory con 
ditions pertain, is also of two kinds, the appetent and non- 
appetent. The appetent is the spirit associated with an 
organism and organs ; the non-appetent is the spirit apart 
from organism and organs. The details of all this are 
to be found in the Panchartha-bhashyadipika and other 
works. The cause is that which retracts into itself and 
evolves the whole creation. This though one is said to 
be divided according to a difference of attributes and 
actions (into Mahesvara, Vishnu, &c.) The Lord is the 
possessor of infinite, visual, and active power. He is 
absolutely first as connected eternally with this lordship 
or supremacy, as possessing a supremacy not adventitious 
or contingent. This is expounded by the author of the 
Adarsa, and other institutional authorities. 

Union is a conjunction of the soul with God through 
the intellect, and is of two degrees, that characterised by 
action, and that characterised by cessation of action. Of 
these, union characterised by action consists of pious 
muttering, meditation, and so forth ; union characterised 
by cessation of action is called consciousness, &c. 


Bite or ritual is activity efficacious of merit as its end. 
It is of two orders, the principal and the subsidiary. Of 
these, the principal is the direct means of merit, religious 
exercise. Religious exercise is of two kinds, acts of piety 
and postures. The acts of piety are bathing with sand, 
lying upon sand, oblations, mutterings, and devotional 
perambulation. Thus the revered JSTakulisa says : 

" He should bathe thrice a day, he should lie upon the 
dust. Oblation is an observance divided into six 

Thus the author of the aphorisms says : 

"He should worship with the six kinds of oblations, 
viz., laughter, song, dance, muttering hum, adora 
tion, and pious ejaculation." 

Laughter is a loud laugh, Aha, Aha, by dilatation of the 
throat and lips. Song is a celebration of the qualities, 
glories, &c., of Mahesvara, according to the conventions of 
the Gandharva-sastra, or art of music. The dance also is 
to be employed according to the ars saltatoria, accompanied 
with gesticulations with hands and feet, and with motions 
of the limbs, and with outward indications of internal 
sentiment. The ejaculation hum is a sacred utterance, 
like the bellowing of a bull, accomplished by a contact 
of the tongue with the palate, an imitation of the sound 
hudung, ascribed to a bull, like the exclamation Vashat. 
Where the uninitiated are, all this should be gone through 
in secret. Other details are too familiar to require ex 

The postures are snoring, trembling, limping, wooing, 
acting absurdly, talking nonsensically. Snoring is showing 
all the signs of being asleep while really awake. Trem 
bling is a convulsive movement of the joints as if under an 
attack of rheumatism. Limping is walking as if the legs 
were disabled. Wooing is simulating the gestures of an 
innamorato on seeing a young and pretty woman. Act 
ing absurdly is doing acts which every one dislikes, as if 
bereft of all sense of what should and what should not 


be done. Talking nonsensically is the utterance of words 
which contradict each other, or which have no meaning 
and the like. 

The subsidiary religious exercise is purificatory sub 
sequent ablution for putting an end to the sense of unfit- 
ness from begging, living on broken food, &c. Thus it is 
said by the author of the aphorisms : Bearing the marks 
of purity by after-bathing. 

(It has been stated above that omniscience, a form of 
the cognitive power, is cognition of principles ever arising 
and pervaded by truth, relative to all matters declared or 
not declared, summary, or in detail). The summary is the 
enouncement of the subjects of attributes generally. This 
is accomplished in the first aphorism: (Now then we 
shall expound the Pasupata union and rites of Pasupati). 
Detail is the fivefold enouncement of the five categories 
according to the instruments of true knowledge. This is 
to be found in the Ptasikara-bhashya. Distribution is the 
distinct enouncement of these categories, as far as possible 
according to definitions. It is an enumeration of these 
according to their prevailing characters, different from 
that of other recognised systems. For example, the cessa 
tion of pain (or emancipation) is in other systems (as in 
the Sankhya) the mere termination of miseries, but in this 
system it is the attainment of supremacy or of the divine 
perfections. In other systems the create is that which 
has become, and that which shall become, but in this 
system it is eternal, the spirits, and so forth, the sentient 
and insentient. In other systems the principium is deter 
mined in its evolution or creative activity by the efficacy 
of works, whereas in this system the principium is the 
Lord not thus determined. In other institutes union re 
sults in isolation, &c., while in these institutes it results 
in cessation of pains by attainment of the divine perfec 
tions. In other systems paradise and similar spheres 
involve a return to metempsychosis, but in this system 
they result in nearness to the Supreme Being, either 


followed or not followed by such return to transmigratory 

Great, indeed, an opponent may say, is this aggregate 
of illusions, since if God s causality be irrespective of the 
efficacy of works, then merits will be fruitless, and all 
created things will be simultaneously evolved (there being 
no reason why this should be created at one time, and that 
at another), and thus there will emerge two difficulties. 
Think not so, replies the Pasupata, for your supposition is 
baseless. If the Lord, irrespective of the efficacy of works, 
be the cause of all, and thus the efficacy of works be with 
out results, what follows ? If you rejoin that an absence 
of motives will follow, in whom, we ask, will this absence 
of motives follow ? If the efficacy of works be without 
result, will causality belong to the doer of the works as to 
the Lord ? It cannot belong to the doer of the works, for 
it is allowed that the efficacy of works is fruitful only 
when furthered by the will of the creator, and the efficacy 
so furthered may sometimes be fruitless, as in the case of 
the works of Yayati, and others. From this it will by 
no means follow that no one will engage in works, for they 
will engage in them as the husbandman engages in hus 
bandry, though the crop be uncertain. Again, sentient 
creatures engage in works because they depend on the 
will of the creator. Nor does the causality pertain to the 
Lord alone, for as all his desires are already satisfied, he 
cannot be actuated by motives to be realised by works. 
As for your statement, continues the Pasupata, that all 
things will be simultaneously evolved, this is unreason 
able, inasmuch as we hold that causal efficiency resides in 
the unobstructed active power which conforms itself to 
the will of the Lord, whose power is inconceivable. It has 
accordingly been said by those versed in sacred tradition: 

" Since he, acting according to his will, is not actuated 
by the efficacy of works, 

" For this reason is he in this system the cause of all 



Some one may urge : In another system emancipation 
is attained through a knowledge of God, where does the 
difference lie ? Say not so, replies the Pasupata, for you 
\vill be caught in a trilemma. Is the mere knowledge of 
God the cause of emancipation, or the presentation, or the 
accurate characterisation, of God ? Not the mere know 
ledge, for then it would follow that the study of any 
system would be superfluous, inasmuch as without any 
institutional system one might, like the uninstructed, 
attain emancipation by the bare cognition that Mahadeva 
is the lord of the gods. Nor is presentation or intuition 
of the deity the cause of emancipation, for no intuition of 
the deity is competent to sentient creatures burdened with 
an accumulation of various impurities, and able to see only 
with the eyes of the flesh. On the third alternative, viz., 
that the cause of emancipation is an accurate characterisa 
tion of the deity, you will be obliged to consent to our 
doctrine, inasmuch as such accurate characterisation can 
not be realised apart from the system of the Pasupatas. 
Therefore it is that our great teacher has said : 

"If by mere knowledge, it is not according to any 
system, but intuition is unattainable ; 

"There is no accurate characterisation of principles 
otherwise than by the five categories." 

Therefore those excellent persons who aspire to the 
highest end of man must adopt the system of the Pasu 
patas, which undertakes the exposition of the five cate 
gories. A. E. G. 



[THE seventh system in Madhava s Sarva-darsana-san- 
graha is the Saiva-darsana. This sect is very prevalent 
in the South of India, especially in tlie Tamil country ; it 
is said to have arisen there about the eleventh century A.D. 
Several valuable contributions have been lately made to 
our knowledge of its tenets in the publications of the Eev. 
H. E. Hoisington and the Eev. T. Foulkes. The former 
especially, by his excellent articles in the American 
Oriental Society s Journal, has performed a great service 
to the students of Hindu philosophy. He has there 
translated the Tattuva-Kattalei, or law of the Tattwas, the 
Siva-Gnanapotham, or instruction in the knowledge of 
God, and the &va-Pirakasam, or light of Siva, and the 
three works shed immense light on the outline as given 
by Madhava. One great use of the latter is to enable us 
to recognise the original Sanskrit names in their Tamil 
disguise, no easy matter occasionally, as arul for anugraha, 
and tidchei for dikshd may testify. 

The Saivas have considerable resemblance to the Theistic 
Sankhya ; they hold that God, souls, and matter are from 
eternity distinct entities, and the object of philosophy is to 
disunite the soul from matter and gradually to unite it to 
God. Siva is the chief deity of the system, and the relation 
between the three is Quaintly expressed by the allegory 
of a beast, its fetters, and its owner. Pasupati is a well- 
known name of Siva, as the master or creator of all things. 


There seem to be three different sets of so-called Saiva 
sutra s. One is in five books, called by Colebrooke the 
Pasupati-sastra, which is probably the work quoted by 
Madhava in his account of the Nakulisa Pasupatas; 
another is in three books, with a commentary by Kshe- 
maraja, with its first sutra, chaitanyam dtmd. The third 
was commented on by Abhinava-gupta, and opens with 
the sloka given in the Sarva-Dargana-Saiigraha, p. 91, lines 
1-4. The MS. which I consulted in Calcutta read the 
first words 

KathancJiid dsddya Maliesvarasya d&syam. 

None of these works, however, appear to be the autho 
rity of the present sect. They seem chiefly to have relied 
on the twenty-eight Agamas and some of the Puranas. 
A list of the Agamas is given in Mr. Foulkes " Catechism 
of the Saiva Keligion;" and of these the Kirana and Karana 
are quoted in the following treatise.] 


Certain, however, of the Mahesvara sect receiving the 
system of truth authoritatively laid down in the Saiva 
Agama, 1 reject the foregoing opinion that "the Supreme 
Being is a cause as independent of our actions, &c.," on the 
ground of its being liable to the imputation of partiality 
and cruelty. They, on the contrary, hold the opinion 
that " the Supreme Being is a cause in dependence on our 
actions, &c.;" and they maintain that there are three cate 
gories distinguished as the Lord, the soul, and the world 
(or literally "the master," "the cattle/ and "the fetter"). 
As has been said by those well versed in the Tantra 

" The Guru of the world, having first condensed in one 

1 Colebrooke speaks of the Pasu- to he twenty-eight (see their names 

pati-sdstra (Mahesvara-siddhdnta or in the Rev. T. Foulkes " Catechism 

Sivdgama), as the text-book of the of the Saiva Religion "). 
Piisupata sect. The Agamas are said 


siitra the great tantra, possessed of three categories 
and four feet, has again declared the same at full 

The meaning of this is as follows : Its three categories 
are the three "before mentioned ; its four feet are learning, 
ceremonial action, meditation, and morality, hence it is 
called the great Tantra, possessed of three categories and 
four feet. Now the " souls " are not independent, and the 
"fetters" are unintelligent, hence the Lord, as being 
different from these, is first declared ; next follows the 
account of the souls as they agree with him in possessing 
intelligence ; lastly follow the " fetters " or matter, such 
is the order of the arrangement. 1 Since the ceremony of 
initiation is the means to the highest human end, and this 
cannot be accomplished without knowledge which estab 
lishes the undoubted greatness of the hymns, the Lords of 
the hymns, &c., and is a means for the ascertainment of 
the real nature of the "cattle," the "fetter," and the 
" master," we place as first the " foot" of knowledge (jndna) 
which makes known all this unto us. 2 Next follows the 
" foot " of ceremonial action (kriya) which declares the 
various rules of initiation with the divers component parts 
thereof. Without meditation the end cannot be attained, 
hence the " foot " of meditation (yoga) follows next, which 
declares the various kinds of yoga with their several parts. 
And as meditation is worthless without practice, i.e., the 
fulfilling what is enjoined and the abstaining from what is 

1 " There must be three eternal 
entities, Deity, soul, matter ; " " as 
the water is co-eternal with the sea 
and the salt with the water, so soul 
is co-eternal with the Deity, and 
pdiia is eternally co-existent with 
soul" (J. A. O. S. iv. pp. 67, 85). 
In p. 58 we find the advaita of the 
Vedaiita attacked. In p. 62 it is 
said that the soul is eternally en 
tangled in matter, and God carries 
on his five operations (see infra) to 
disentangle it, bringing out all that 
is required for previous desert. 

2 These four feet are the four 
stages of religious life (see J. A. O. S. 
iv. pp. 135, 1 80), called in Tamil 
sarithei, Jcirikei, yoJcam, and gndnam. 
The first is the stage of practical 
piety and performance of the pre 
scribed duties and rites ; the second 
is that of the " confirmatory sacra 
ment " and the five purifications in 
volved in true pujd ; the third is 
that of the eight observances of the 
yogin ; the fourth is that of know 
ledge which prepares the soul for 
intimate union with God. 


forbidden, lastly follows the fourth "foot" of practical 
duty (charyd\ which includes all this. 

Now Siva is held to be the Lord (or master). Although, 
participation in the divine nature of Siva belongs to 
liberated souls and to such beings as Vidye^vara, &c., yet 
these are not independent, since they depend on the 
Supreme Being; and the nature of an effect is recognised 
to belong to the worlds, &c., which resemble him, from the 
very fact of the orderly arrangement of their parts. And 
from their thus being effects we infer that they must have 
been caused by an intelligent being. By the strength of 
this inference is the universal acknowledgment of a 
Supreme Being confirmed. 

" But may we not object that it is not proved that the 
body is thus an effect ? for certainly none has ever, at any 
time or place, seen a body being made by any one." We 
grant it : yet it is not proper to deny that a body has some 
maker on the ground that its being made has not been 
seen by any one, since this can be established from infer 
ence [if not from actual perception]. Bodies, &c., must 
be effects, because they possess an orderly arrangement of 
parts, or because they are destructible, as jars, &c. ; and 
from their being effects it is easy to infer that they must 
have been caused by an intelligent being. Thus the sub 
ject in the argument [sc. bodies, &c.] must have had a 
maker, from the fact that it is an effect, like jars, &c. ; that 
which has the aforementioned middle term (sddkana) must 
have the aforementioned major (sadhyct) ; and that which 
has not the former will not have the latter, as the soul, 
&C. 1 The argument which establishes the authority of 
the original inference to prove a Supreme Being has been 
given elsewhere, so we refrain from giving it at length 
here. In fact, that God is the universal agent, but not 
irrespective of the actions done by living beings, is proved 
by the current verse 2 

1 Cf. Colebrooke, Essays (26. ed.), vol. i. p. 315. 

2 Nydyena may here mean " argument." 


" This ignorant jivdtman, incapable of its own true 
pleasures or pains, if it were only under God s direc 
tion [and its own merits not taken into account], 
would always go to heaven or always to hell." * 
Nor can you object that this opinion violates God s 
independence, since it does not really violate an agent s 
independence to allow that he does not act irrespectively 
of means ; just as we say that the king s bounty shows 
itself in gifts, but these are not irrespective of his trea 
surer. As has been said by the Siddha Guru 

"It belongs to independence to be uncontrolled and 

itself to employ means, &c. ; 

" This is an agent s true independence, and not the act 
ing irrespectively of works, &c." 

And thus we conclude that inference (as well as Sruti) 
establishes the existence of an agent who knows the various 
fruits [of action], their means, material causes, &c., accord 
ing to the laws of the various individual merits. This has 
been thus declared by the venerable Brihaspati 

" He who knows the fruits to be enjoyed, their means 

and material causes, 

" Apart from him this world knows not how the desert 
that resides in accumulated actions should ripen." 
" The universe is the subject of our argument, and it 
must have had an intelligent maker. 

O * 

" This we maintain from its being an effect, just as we 
see in any other effect, as jars, &c." 

God s omniscience also is proved from his being identical 
with everything, and also from the fact that an ignorant 
being cannot produce a thing. 2 This has been said by the 
illustrious Mrigendra 3 

1 Scil. if there were only one cause meaning of the passage ; it occurs 
there would be only one invariable Mahjibharata, iii. 1144 (cf. Gauda- 
effect. The very existence of various pada, S. Kar. 61). 

effects proves that there must be 
other concurrent causes (as human 
actions) necessary. The argument 
seems to me to require here this 
unnatural stress to be laid on era, 
but this is certainly not the original 

2 In p. 82, line 3, infra, I read 

3 This may be the same with the 
Meykanda of the Tamil work in 
J. A. O. S. His poem was called 
the Mrifjendra, (?). 


"He is omniscient from his being the maker of all 
things : for it is an established principle 

" That he only can make a thing who knows it with its 
means, parts, and end." 

" Well," our opponents may say, " we concede that God 
is an independent maker, but then he has no body. 1 
Now experience shows that all effects, as jars, &c., are 
produced by beings possessed of bodies, as potters, &c. ; 
but if God were possessed of a body, then he would be 
like us subject to trouble, and no longer be omniscient or 
omnipotent." We, however, deny this, for we see that 
the incorporeal soul does still produce motion, &c., in its 
associated body ; moreover, even though we conceded that 
God did possess a body, we should still maintain that the 
alleged defects would not necessarily ensue. The Supreme 
Being, as he has no possible connection with the fetters 
of matter, such as mala, 2 action, &c., cannot have a 
material body, but only a body of pure energy (Sakta), 3 
since we know that his body is composed of the five 
hymns which are forms of Sakti, according to the well- 
known text : " The Supreme has the Isdna as his head, 
the Tatpurusha as his mouth, the Aghora as his heart, the 
Vdmedeva as his secret parts, and the Sadyojdta, as his 
feet." 4 And this body, created according to his own will, 
is not like our bodies, but is the cause of the five opera 
tions of the Supreme, which are respectively grace, obscura 
tion, destruction, preservation, and production. 5 This has 
been said in the Srimat Mrigendra 

1 Should we read tdvad a nasarirah of Siva (see J. A. O. S. iv. p. 101). 

in p. 83, line 2 ? These five mantras are given in the 

- I retain this word, see infra. inverse order in Taitt. A ranyaka, x. 

3 "Maya (orPrakriti) is the mate- 43-47 (cf. Nyayd-mdldvist. p. 3). 
rial, Sakti the instrumental, and 5 These are the operations of the 
Deity the efficient cause " ^ J. A. 0. S. five manifestations of Siva (see 
iv. p. 55). J. A. O. S. iv. 8, 1 8) which in their 

4 These are the five first names of descending order are SdthdJckiyam 
the eleven mantras which are in- (i.e., Saddkshaya h or Sadd-S ira, who 
eluded in the five kalds (J. A. O. S. is Siva and Sakti combined, and the 
iv. pp. 238-243). The Sivalinga (the source of grace to all souls ; Ichchuran 
visible object of worship for the en- or Mayesuran, the obscure ; Sutta- 
lightened) is composed of mantras, vittei (8 uddharidyd) which is pro- 
and is to be regarded as the body perly the Hindu triad, Rudra, Vishnu, 


" From the impossibility of its possessing mala, &c., the 
body of the Supreme is of pure energy, and not 
like ours." 
And it has also been said elsewhere 

" His body is composed of the five mantras which are 

subservient to the five operations, 
" And his head, &c., are formed out of the f sa, Tatpur- 

usha, Aghora, Varna, and other hymns." 
If you object to this view that " such passages in the 
Agamas as He is five-faced and fifteen-eyed, assert pro 
minently the fact that the Supreme Being is endowed 
with a body, organs, &c.," we concede what you say, but 
we maintain that there is no contradiction in his assuming 
such forms to show his mercy to his devoted servants, 
since meditation, worship, &c., are impossible towards a 
Being entirely destitute of form. This has been said in 
the Paushkara 

" This form of his is mentioned for the preservation of 

the devotee." 
And similarly elsewhere 

l< Thou art to be worshipped according to rule as pos 
sessed of form ; 
" Tor the understanding cannot reach to a formless 


Bhojaraja 1 has thus detailed the five operations 
"Fivefold are his operations, creation, preservation, 

destruction, and obscuration, 
"And to these must be added the active grace of him 

who is eternally exalted." 

Now these five operations, in the view of the pure Path, 
are held to be performed directly by Siva, but in that of 
the toilsome Path they are ascribed to Ananta, 2 as is 
declared in the Srimat Karana 3 

and Brahma. They are respectively 2 Ananta is a name of Siva in the 

symbolised by the ndda, vindu, m, Atharva-siras Upanishad (see In- 

u, and a of Om. dische Stud. i. 385). 

1 In Wilson s Mackenzie Cat. i. 3 This is the fourth of the twenty - 

p. 138, we find a Tantrik work, the eight Agamas (see Foulkes Cate- 

Narapati-jaya-charyd, ascribed to chism). 
Bhoja the king of DMr. 


"In the Pure Path Siva is declared to be the only 
agent, but Ananta in that which is opposed to the 
One Supreme." 

It must here be understood that the word Siva includes 
in its proper meaning "the Lord," all those who have 
attained to the state of Siva, as the Lords of the Mantras, 
Maheswara, the emancipated souls who have become Sivas, 
and the inspired teachers (vdchakas), together with all the 
various means, as initiation, &c., for obtaining the state of 
Siva. Thus has been explained the first category, the 
Lord (pati). 

We now proceed to explain the second category, the 
soul (pasu). The individual soul which is also known by 
such synonyms as the non-atomic, 1 the Kshetrajna, or 
knower of the body, 2 &c., is the Pasu. For we must not 
say with the Charvakas that it is the same as the body, 
since on this view we could not account for memory, as 
there is a proverb that one man cannot remember what 
another has seen. Nor may we say with the Naiyayikas 
that it is cognisable by perception, 3 as this would involve 
an ad infinitum regressus. As has been said 

" If the soul were cognisable, there would need to be 
again a second knower ; 4 

"And this would require another still, if the second 
were itself to be known." 

Nor must we hold it non-pervading with the Jainas, 
nor momentary with the Bauddhas, since it is not limited 
by space or time. As has been said 

" That object which is unlimited in its nature by space 
or time, 

"They hold to be eternal and pervading, hence the 
soul s all-pervadingness and eternity." 

1 Ann ? "The soul, when clothed 2 See Ind. Studien, i. 301. 

with these primary things (desire, 3 The mind or internal sense per- 

knowledge, action, &c. ), is an exceed- ceives soul (see Bhdsha" Parich- 

ingly small body" (Foulkes). Ananu chheda, sloka 49). 

is used as an epithet of Brahman in 4 Dele the iti in p. 84, line 5, 

Brihad Ar. Up. iii. 8. 8. infra. 


Nor may we say with the Vedantin that it is only one, 
since the apportionment of different fruits proves that 
there are many individual souls ; nor with the Sankhyas 
that it is devoid of action, since, when all the various 
" fetters " are removed, Sruti informs us of a state of 
identity with Siva, which consists in intelligence in the 
form of an eternal and infinite vision and action. 1 This 
has been declared in the Srirnat Mrigendra 

" It is revealed that identity with Siva results when all 

fetters are removed." 
And again 

" Intelligence consists in vision and action, and since in 

his soul 

" This exists always and on every side, therefore, after 
liberation, Sruti calls it that which faces every 
It is also said in the Tattva-prakasa 

" The liberated souls are themselves Sivas, but these 

are liberated by his favour; 
"He is to be known as the one eternally liberated, 

whose body is the five Mantras." 

Now the souls are threefold, as denominated vijftdnd- 
Jcaldh, pralaydkaldh, and sakaldh. 2 (.) The -first are those 
who are under the influence of mala only, since their 
actions are cancelled by receiving their proper fruits, or 

^ Of. the Nakulisa Pasupatas, p. where it is said that the five vidyd- 

76, 4 (supra, p. 103). tattvas (kald, mdyd, rdga, nu/ati, and 

2 For these three classes see kald) and the twenty-four dtrna- 

J. A. O. S. iv. pp. 87, 137. They tattvas (sc. the gross and subtile 

are there described as being respec- elements, and organs of sense and 

tively under the influence of dnavam action, with the intellectual faculties 

malam only, or this with kanmam manas, buddhi, ahamkdra, and chitta), 

malam, or these with mayei malam. are all developed from mdyd. This 

The dnavam is described as original exactly agrees with the quotation 

sin, or that source of evil which was from Soma Sambhu, infra. We may 

always attached to the soul ; kan- compare with it what Madhava says, 

mam is that fate which inheres in p. 77, in his account of the Nakulisa 

the soul s organism and metes out P&supatas, where he describes kald 

its deserts ; mayei is matter in its as unintelligent, and composed of 

obscuring or entangling power, the the five elements, the five tanmdtras, 

source of the senses. Mddhava uses and the ten organs, with buddhi, 

"kald" &c., for mdyd. The reason ahamkdra and manas. 
is to be found in J. A. O. S. p. 70, 


by abstraction, contemplation, and knowledge, and since 
they have no " fetters " in the form of enjoyments, such 
as kald, &c. (which fetters would, however, be the cause of 
cancelling actions by bringing about their proper fruit). 
(5.) The second are those who are under the influence of 
mala and barman, since in their case kald, &c., are de 
stroyed by mundane destructions, hence their name prala- 
ydkala. (c.) The third are those who are bound in the 
three fetters of mala, mdyd, and barman, hence their name 
saJcala. The first class are again subdivided into samdpta- 
kalushdh and asamdpta-kalushdh, according as their in 
herent corruption is perfectly exhausted or not. The 
former, having received the mature penalties of their 
corruptions, are now, as foremost of men and worthy of 
the privilege, raised by Siva s favour to the rank of the 
Lords of Knowledge (the Vidyesvaras), Ananta, and the 
rest. This ogdoad of the Lords of Knowledge is described 
in the Bahudaivatya 

" Ananta, and Sukshma, and Sivottama, 

" Ekanetra, and again Ekarudra and Trimurttika, 

" Srikantha and Sikhandin, these are declared to be 
the Vidyesvaras." 

The latter Siva, in his mercy, raises to the rank of the 
seventy million Mantras. 1 All this is explained in the 
Tattva-prakasa. 2 Similarly Soma-Sambhu has said 

"One class is named vijndndkala, the second prala- 

"The third sakala, these are the three whom the 
Sastra regards as objects of mercy. 

" The first is united to mala alone, the second to mala 
and karma, 

" The third are united to all the tattvas beginning with 
kald and ending with " earth." 3 

1 See J. A. 0. S. iv. p. 137. I read vijndna-Jcevala, pralaya-Tcevala, and 
anugrahakarandt in p. 86, line 3. sakala. 

2 I omit the quotation, as it only 3 I.e., thus including five of the 
repeats the preceding. It, how- vidydtattras and all the twenty-four 
ever, names the three classes as dtmatattvas. 


The Pralaydkaldh are also twofold, as being pakvapdsa- 
dvaija or not, i.e., those in whom the two remaining fetters 
are matured, and those in whom they are not. The 
former attain liberation, but the latter, by the power of 
Jcarman, are endowed with the puryashtaka^ body, and 
pass through various births. As has been said in the 

" Those among the Pralayakalas whose karman and mala 
are immature, 

"Go, united with the puryashtaka body, into many 
births by the power of karman" 

Thepuryashtakais also thus described in the same work 

"The puryashtaka is composed of the internal organ, 
thought (dhi), barman, and the instruments." 

This is thus explained by Aghora Siva Acharya, " the 
puryasJitaka is a subtile body apportioned to each indi 
vidual soul, which continues from the creation until the 
close of the kalpa, or until liberation : it is composed of 
the thirty 2 tattvas beginning with earth and ending 
with Jcald" As has been said in the Tattva-sangraha 

" This set of tattvas, commencing with earth and end 
ing with kald, is assigned to each soul, 

" And wanders by the law of karman through all the 
bodies produced by the world." 

The following is the full meaning of this passage : 
The word " internal organ! which properly includes 
"mind," "intelligence," "egoism," and "reason," 3 includes 
also the seven tattvas which enter into the production of 
enjoyment [or experience], viz., those called kald, time, 
fate, knowledge, concupiscence, nature, and quality ; 4 the 

1 This term seems to be derived prakriti, and guna. Hoisington, how- 

from puri, "body (cf. purisaya for ever, puts purushan "the principle 

purusha, Brihad Ar. Up. ii. 5, 18), of life," instead of guna, which seems 

and ashtaka (cf. also the Sdnkhya better, as the threegunas are included 

Pravachana Bhftshya, p. 135). in prakriti. He translates kald by 

2 Or rather thirty-one ? " continency," and describes it as 

3 Manas, buddhi, ahamkdra, chitta. " the power by which the senses are 

4 These are the seven vid yd tattvas, subdued and the carnal self brought 
bald, kola, niyati (fate), vidyd, rdga, into subjection." 


words " tlwught (dhi) and karman signify the five cog 
nisable gross elements, and their originators, the subtile 
rudiments. By the word " instruments " are comprehended 
the ten organs of sense and action. 

" But is it not declared in the Srimat Kalottara that 
The set of five, sound, touch, form, taste and smell, in 
telligence, mind and egoism, these constitute the pur- 
yaslitaka ? " 

How, then, can any different account be maintained ? 
We grant this, and hence the venerable Bama Kantha has 
explained that sutra in its literal meaning [i.e, as puryash- 
taJea, is derived from ashta, " eight "], so why should we 
be prolix in the discussion ? Still, if you ask how we can 
reconcile our account with the strict nominal definition of 
puryashtaka, we reply that there is really no contradiction, 
as we maintain that it is composed of a set of eight in the 
following manner: (i.) The five elements; (2.) the five 
rudiments; (3.) the five organs of knowledge; (4.) those 
of action; (5.) the fourfold internal organ; (6.) their in 
strument; 1 (7.) nature [prakriti] ; and (8.) the class com 
posed of the five, beginning with kald, which form a kind 
of case. 2 

Now in the case of some of those souls who are joined 
to the puryashtaka body, Mahesvara Ananta having com 
passionated them as possessed of peculiar merit, constitutes 
them here as lords of the world ; as has been said 

" Mahesvara pities some and grants them to be lords of 
the world." 

The class called saJcala is also divided into two, as 
pakvakalusha and apakvakalusha. As for the former, the 
Supreme Being, in conformity with their maturity (pari- 

1 This " instrument " (karana) 2 The thirty-one tattvas are as 

seems to mean what Hoisington calls follow : Twenty-four dtma-tattvas, 

purushan or "the principle of life five elements, five tanmdtras, ten 

which establishes or supports the organs of sense and action, four 

whole system in its operation ; " he organs of the antalikarana, and seven 

makes it one of the seven vidyd- mdydtattvas as enumerated above. 

tattvas. According to Madhava, it (See J. A. O. S. iv. pp. 16-17.) 
should be what he calls guna. 


pdka), puts forth a power agreeable thereto, and transfers 
them to the position of the hundred and eighteen Lords of 
the Mantras, signified by the words Mandali, &c., as has 
been said 

" The rest are denominated sakala, from their connection 
with Kald, &c., seized by time whose mouths are 
days ; 
" The Supreme of his own will makes one hundred and 

eighteen of these the Lords of the Mantras. 
" Eight of these are called Mandolins ; eight again are 

Krodha, &c. ; 
"Viresa, Srikantha, and the hundred Rudras, these 

together are the hundred and eighteen." 
In their case again, the Supreme, having assumed the 
form of a teacher, stops the continued accession of maturity 
and contracts his manifested power, and ultimately grants 
to them liberation by the process of initiation; as has 
been said 

" These creatures whose mala is matured, by putting 

forth a healing power, 

" He, assuming the form of a teacher, unites by initia 
tion to the highest principle." 
It is also said in the Srimad Mrigendra 

" He removes from that infinitesimal soul all the bonds 
which previously exerted a contrary influence over 
it." 1 

All this has been explained at great length by ]Nara- 
yana-Kantha, and there it is to be studied; but we are 
obliged to pass on through fear of prolixity. 

But as for the second class, or those called apakmka- 
lusha, the Supreme Being, as impelled by the desert of 
their respective actions, appoints them, as bound and 
endued with infinitesimal bodies, to enjoy the rewards of 
their previous actions. 2 As has been said 

1 I take ami in this verse as the mdyd-mala, the second dnava-mala, 

soul, but it may mean the second the third kanma-mala (karman). 
kind of mala mentioned by Hoising- 2 " The soul, when clothed with 

ton. The first kind of mala is the these primary things (desire, know- 


" The other souls, bound [in their material bonds] he 
appoints to enjoy their various deserts, 

According to their respective actions: such are the 
various kinds of souls." 

We now proceed to describe the third category, matter 
(or pdsa). This is fourfold, mala, 1 karman, mdyd, and 
rodha-sakti. 1 But it may be objected, " Is it not said in 
the Saiva Agamas that the chief things are the Lord, souls, 
and matter? Now the Lord has been shown to mean 
Siva, souls mean atoms (or beings endowed with atomic 
bodies), and matter (or bond ) is said to be the pentad, 2 
hence matter will be fivefold. How then is it now 
reckoned to be only fourfold ? " To this we reply as 
follows : Although the vindu or nasal dot, which is the 
germinal atom of mdyd, and is called a Siva-tattva, 
may be well regarded as material in comparison with 
the highest liberation as defined by the attainment of 
the state of Siva, still it cannot really be considered 
as matter when we remember that it is a secondary 
kind of liberation as causing the attainment of the 
state of such deities as Vidyesvara, &c. Thus we see 

ledge, action, the Tcalddipancliaka, developed. From this atom are 

&c.), is an exceedingly small body " developed the four sounds, the fifty- 

(Foulkes). One of the three malas one Sanskrit letters, the Vedas, 

is called dnara, and is described as Mantras, &c., the bodily, intellec- 

the source of sin and suffering to tual, and external enjoyments of 

souls. the soul that have not attained to 

1 The first three are the three spiritual knowledge at the end of 
kinds of mala in the J. A. O. S., viz., each period of the world s existence, 
dnavam, kanmam, and mdyei, the last and have been swept away by the 
is the " obscuring " powfer of Mdye- waters of the world - destroying 
suran (cf. vol. iv. pp. 13, 14. The deluge; after these the three stages 
Saivas hold that Pasa, like the Sdn- of heavenly happiness are developed, 
khya Prakriti, is in itself eternal, to be enjoyed by the souls that have 
although its connection with any a favourable balance of meritorious 
particular soul is temporary (see deeds, or have devoted themselves 
J. A. 0. S. iv. p. 228). to the service of God or the abstract 

2 These are the five, vindu, mala, contemplation of the Deity, viz., 
karman,mdyd,a,ndrodhasakti. Vindu (i.) the enjoyment of the abode of 
is described in Foulkes translation Siva ; (2.) that of near approach to 
of the Siva-prakasa-patalai : "A him; (3.) that of union with him." 
sound proceeds out of the mystical Vindu is similarly described, J. A. 
syllable om ; . . . and in that sound O. S. iv. pp. 152, 153 (cf. also Weber, 
a rudimentary atom of matter is RdmatApanyia Up. pp. 312-315). 


there is no contradiction. Hence it has been said in the 

" The bonds of matter will be fourfold." 
And again in the Srimad Mrigendra 

" The enveloper-controller (mala), the overpowerer 
(rodha), action, and the work of Maya, 

" These are the four bonds, and they are collectively 
called by the name of merit. " 

The following is the meaning of this couplet : 

(i.) "Enveloping," because mala exceedingly obscures 
and veils the soul s powers of vision and action ; " con 
trolling," because mala, a natural impurity, controls the 
soul by its independent influence. As has been said 

" Mala, though itself one, by manifold influence inter 
rupts the soul s vision and action ; 

"It is to be regarded as the husk in rice or rust on copper." 1 

(2.) The " overpowerer " is the obscuring power ; this is 
called a " bond " [or matter] in a metaphorical sense, since 
this energy of Siva obscures the soul by superintending 
matter [rather than by itself partaking of the nature of 

Thus it has been said 

" Of these I am the chief energy, and the gracious friend 
of all, 

" I am metaphorically called pdsa, 2 because I follow 

(3.) Action [or rather its consequences, Jcarman] as 
being performed by those who desire the fruit. It is in 
the form of merit or demerit, like the seed and shoot, and 
it is eternal in a never-beginning series. As has been 
said in the Srimat Kirana 

" As Mala has no beginning, its least actions are begin- 
ningless : 

" If an eternal character is thus established, then what 
cause could produce any change therein ? " 

1 See the same illustrations in J. A. O. S. iv. p. 150. 
2 Some forced derivation seems here intended as of pdaa from pascMt. 


(4.) " Mdyd," because herein as an energy of the Divine 
Being all the world is potentially contained (mdti) at a 
mundane destruction, and again at a creation it all comes 
(ydti) into manifestation, hence the derivation of the 
name. This has been said in the Srimat Saurabheya 
" The effects, as a form of the Divine energy, are absorbed 

therein at a mundane destruction, 
" And again at a renovation it is manifested anew in the 

form of effects as hold, &c." l 

Although much more might be added on this topic, yet 
we stop here through fear of extending this treatise too 
far. Thus have the three categories been declared, the 
Lord, the soul, and matter. 

A different mode of treating the subject is found in the 
Jnanaratnavali, &c., in such lines as 

" The Lord, knowledge, ignorance, the soul, matter, and 

the cause 
"Of the cessation thereof, these are collectively the 

six categories." 

But our readers must seek for full information from the 
work itself. Thus our account of the system is complete. 

E. B. C. 

1 In p. 90, line 2, read sd Mrycna. 

( 128 ) 



OTHER Mahesvaras are dissatisfied with the views set out 
in the Saiva system as erroneous in attributing to motive 
less and insentient things causality (in regard to the bond 
age and liberation of transmigrating spirits). They there 
fore seek another system, and proclaim that the construction 
of the world (or series of environments of those spirits) is 
by the mere will of the Supreme Lord. They pronounce 
that this Supreme Lord, who is at once other than and the 
same with the several cognitions and cognita, who is 
identical with the transcendent self posited by one s own 
consciousness, by rational proof, and by revelation, and 
who possesses independence, that is, the power of witness 
ing all things without reference to aught ulterior, gives 
manifestation, in the mirror of one s own soul, to all 
entities l as if they were images reflected upon it. Thus 
looking upon recognition as a new method for the attain 
ment of ends and of the highest end, available to all men 
alike, without any the slightest trouble and exertion, such 
as external and internal worship, suppression of the breath, 
and the like, these Mahesvaras set forth the system of 
recognition (pratyabhijnd). The extent of this system is 
thus described by one of their authorities 

"The aphorisms, the commentary, the gloss, the two 
explications, the greater and the less, 

1 Read Widvdn for bhdvdt. 


"The five topics, and the expositions, such is the 
system of recognition." 

The first aphorism in their text-book is as follows l : 

" Having reached somehow or other the condition of a 
slave of Mahesvara, and wishing also to help man 

" I set forth the recognition of Mahesvara, as the method 
of attaining all felicity." 

[This aphorism may be developed as follows] : 

" Somehow or other," by a propitiation, effected by God, 
of the lotus feet of a spiritual director identical with God, 
"having reached," having fully attained, this condition, hav 
ing made it the tmintercepted object of fruition to myself. 
Thus knowing that which has to be known, he is qualified 
to construct a system for others: otherwise the system 
would be a mere imposture. 

Mahesvara is the reality of unintermitted self-luminous- 
ness, beatitude, and independence, by portions of whose 
divine essence Vishnu, Virinchi, and other deities are 
deities, who, though they transcend the fictitious world, 
are yet implicated in the infinite illusion. 

The condition of being a slave to Mahesvara is the being 
a recipient of that independence or absoluteness which is 
the essence of the divine nature, a slave being one to 
whom his lord grants all things according to his will and 
pleasure (i.e., ddsya, from da). 

The word mankind imports that there is no restriction 
of the doctrine to previously qualified students. Whoever 
he may be to whom this exposition of the divine nature is 
made, he reaps its highest reward, the Qmauatoiy principium 
itself operating to the highest end of the transmigrating 
souls. It has been accordingly laid down in the Siva- 
drishti by that supreme guide the revered Somananda- 

" When once the nature of Siva that resides in all things 

1 Cf. supra, p. 113. Madhava in the beginning of the eleventh 
here condenses Abhinava Gupta s century (see Blihler s Tour in Cash- 
commentary. Abhinava Gupta lived mere, pp. 66, So;. 



has been known with tenacious recognition, whether 

by proof or by instruction in the words of a spiritual 

"There is no further need of doing aught, or of any 

further reflection. When he knows Suvarna (or 

Siva) a man may cease to act and to reflect." 
The word also excludes the supposition that there is 
room in self which has recognised the nature of MahesVara, 
and which manifests to itself its own identity with him, 
and is therefore fully satisfied, for any other motive than 
felicity for others. The well-being of others is a motive, 
whatever may be said, for the definition of a motive applies 
to it : for there is no such divine curse laid upon man that 
self-regard should be his sole motive to the exclusion of a 
regard for others. Thus Akshapada (i. 24) defines a motive : 
A motive is that object towards which a man energises. 

The preposition upa in upapddayami (I set forth) in 
dicates proximity : the result is the bringing of mankind 
near unto God. 

Hence the word all in the phrase the method of attaining 
all felicities. For when the nature of the Supreme Being 
is attained, all felicities, which are but the efflux thereof, 
are overtaken, as if a man acquired the mountain Eohana 
(Adam s Peak), he would acquire all the treasures it con 
tains. If a man acquire the divine nature, what else is 
there that he can ask for? Accordingly Utpalacharya 

" What more can they ask who are rich in the wealth 

of devotion ? What else can they ask who are 

poor in this ? " 

We have thus explained the motive expressed in the 
words the method of attaining all felicities, on the supposi 
tion that the compound term is a Tat-purusha genitively 
constructed. Let it be taken as a Bahuvrihi or relative 
compound. Then the recognition of Mahesvara, the know 
ing him through vicarious idols, has for its motive the full 
attainment, the manifestation, of all felicities, of every 


external and internal permanent happiness in their proper 
nature. In the language of everyday life, recognition is 
a cognition relative to an object represented in memory : 
for example, This (perceived) is the same (as the remem 
bered) Chaitra. In the recognition propounded in this 
system, there being a God whose omnipotence is learnt 
from the accredited legendaries, from accepted revelation, 
and from argumentation, there arises in relation to my 
presented personal self the cognition that I am that very 
God, in virtue of my recollection of the powers of that 

This same recognition I set forth. To set forth is to 
enforce. I establish this recognition by a stringent pro 
cess which renders it convincing. [Such is the articulate 
development of the first aphorism of the Eecognitive 

Here it may be asked : If soul is manifested only as 
consubstantial with God, why this laboured effort to 
exhibit the recognition ? The answer is this : The recog 
nition is thus exhibited, because though the soul is, as 
you contend, continually manifested as self-luminous (and 
therefore identical with God), it is nevertheless under 
the influence of the cosmothetic illusion manifested as 
partial, and therefore the recognition must be exhibited 
by an expansion of the cognitive and active powers in 
order to achieve the manifestation of the soul as total 
(the self being to the natural man a part, to the man of 
insight the whole, of the divine pleroma). Thus, then, the 
syllogism: This self must be God, because it possesses 
cognitive and active powers ; for so far forth as any one 
is cognitive and active, to that extent he is a lord, like a 
lord in the world of everyday life, or like a king, therefore 
the soul is God. The five-membered syllogism is here 
employed, because so long as we deal with the illusory 
order of things, the teaching of the Naiyayikas may be 
accepted. It has thus been said by the son of Udayakara 

" What self-luminous self can affirm or deny that self- 


active and cognitive is MahesVara the primal 
being ? 

" Such recognition must be effected by an expansion of 
the powers, the self being cognised under illusion, 
and imperfectly discerned." 
And again 

" The continuance of all living creatures in this trans- 
migratory world lasts as long as their respiratory 
inwlucrum ; knowledge and action are accounted 
the life of living creatures. 
" Of these, knowledge is spontaneously developed, and 

action (or ritual), which is best at Kasi, 
"Is indicated by others also: different from these is 

real knowledge." 
And also 

" The knowledge of these things follows the sequence 

of those things : 
" The knower, whose essence is beatitude and knowledge 

without succession, is Mahesvara." 
Somananda-natha also says 

" He always knows by identity with Siva : he always 

knows by identity with the real." 
Again at the end of the section on knowledge 

"Unless there were this unity with Siva, cognitions 

could not exist as facts of daily life : 
" Unity with God is proved by the unity of light. He 

is the one knower (or illuminator of cognitions). 
"He is Mahesvara, the great Lord, by reason of the 

unbroken continuity of objects : 
" Pure knowledge and action are the playful activity of 

the deity." 

The following is an explanation of Abhinava-gupta : 
The text, " After that as it shines shines the all of things, 
by the light of that shines diversely this ALL," teaches 
that God illumines the whole round of things by the 
glory of His luminous intelligence, and that the diver 
sity or plurality of the object world, whereby the light 


which irradiates objects is a blue, a yellow light, and the 
like, arises from diversity of tint cast upon the light by the 
object. In reality, God is without plurality or difference, 
as transcending all limitations of space, time, and figure. 
He is pure intelligence, self-luminousness, the manifester ; 
and thus we may read in the Saiva aphorisms, " Self is 
intelligence." His synonymous titles are Intelligential 
Essence, Unintermitted Cognition, Irrespective Intuition, 
Existence as a mass of Beatitude, Supreme Domination. 
This self-same existing self is knowledge. 

By pure knowledge and action (in the passage of Soma- 
nandanatha cited above) are meant real or transcendent 
cognition and activity. Of these, the cognition is self- 
luminousness, the activity is energy constructive of the 
world or series of spheres of transmigratory experience. 
This is described in the section on activity 

" He by his power of bliss gives light unto these objects, 
through the efficacy of his will : this activity is 
And at the close of the same section 

" The mere will of God, when he wills to become the 
world under its forms of jar, of cloth, and other 
objects, is his activity worked out by motive and 

" This process of essence into emanation, whereby if this 
be that comes to be, cannot be attributed to motive 
less, insentient things." 

According to these principles, causality not pertaining 
either to the insentient or to the non-divine intelligence, 
the mere will of Mahesvara, the absolute Lord, when he 
wills to emanate into thousands of forms, as this or that 
difference, this or that action, this or that modification of 
entity, of birth, continuance, and the like, in the series of 
transmigratory environments, his mere will is his pro 
gressively higher and higher activity, that is to say, his 
universal creativeness. 


How he creates the world by his will alone is clearly 
exhibited in the following illustration 

" The tree or jar produced by the mere will of thau- 
maturgists, without clay, without seed, continues 
to serve its proper purpose as tree or jar." 

If clay and similar materials were really the substantial 
cause of the jar and the rest, how could they be produced 
by the mere volition of the thaumaturgist ? If you say : 
Some jars and some plants are made of clay, and spring 
from seeds, while others arise from the bare volition of the 
thaumaturgist ; then we should inform you that it is a 
fact notorious to all the world that different things must 
emanate from different materials. 

As for those who say that a jar or the like cannot be 
made without materials to make it of, and that when a 
thaumaturgist makes one he does so by putting atoms in 
motion by his will, and so composing it: they may be 
informed that unless there is to be a palpable violation of 
the causal relation, all the co-efficients, without exception, 
must be desiderated ; to make the jar there must be the 
clay, the potter s staff, the potter s wheel, and all the rest 
of it ; to make a body there must be the congress of the 
male and female, and the successive results of that con 
gress. Now, if that be the case, the genesis of a jar, a 
body, or the like, upon the mere volition of the thau 
maturgist, would be hardly possible. 

On the other hand, there is no difficulty in supposing 
that Mahadeva, amply free to remain within or to over 
step any limit whatever, the Lord, manifold in his oper- 
ancy, the intelligent principle, thus operates. Thus it is 
that Vasuguptacharya says 

" To him that painted this world-picture without 
materials, without appliances, without a wall to paint it 
on, to him be glory, to him resplendent with the lunar 
digit, to him that bears the trident." 

It may be asked : If the supersensible self be no other 


than God, how comes this implication in successive trans- 
migratory conditions ? The answer is given in the section 
treating of accredited institution 

" This agent of cognition, blinded by illusion, trans 
migrates through the fatality of works : 
" Taught his divine nature by science, as pure intelli 
gence, he is enfranchised." 

CD * 

It may be asked: If the subject and the object are 
identical, what difference can there be between the self 
bound and the self liberated in regard to the objects 
cognisable by each ? The answer to this question is given 
in a section of the Tattvartha-Sangraha 

" Self liberated cognises all that is cognisable as identical 
with itself, like Mahesvara free from bondage : 
the other (or unliberated) self has in it infinite 

An objection may be raised: If the divine nature is 
essential to the soul, there can be no occasion to seek for 
this recognition ; for if all requisites be supplied, the seed 
does not fail to germinate because it is unrecognised. 
Why, then, this toilsome effort for the recognition of the 
soul ? To such an objection we reply : Only listen to the 
secret we shall tell you. All activity about objects is of 
two degrees, being either external, as the activity of the 
seed in developing the plant, or internal, as the activity 
which determines felicity, which consists in an intuition 
which terminates in the conscious self. The first degree 
of activity presupposes no such recognition as the system 
proposes, the second does presuppose it. In the Recogni- 
tive System the peculiar activity is the exertion of the 
power of unifying personal and impersonal spirit, a power 
which is the attainment of the highest and of mediate 
ends, the activity consisting in the intuition I am God. 
To this activity a recognition of the essential nature of 
the soul is a pre-requisite. 

It may be urged that peculiar activity terminating 
in the conscious self is observed independent of recog- 


nition. To this it is replied : A certain damsel, hearing 
of the many good qualities of a particular gallant, fell in 
love with him before she had seen him, and agitated by 
her passion and unable to suffer the pain of not seeing 
him, wrote to him a love-letter descriptive of her condition. 
He at once came to her, but when she saw him she did 
not recognise in him the qualities she had heard about ; 
he appeared much the same as any other man, and she 
found no gratification in his society. So soon, however, as 
she recognised those qualities in him as her companions 
now pointed them out, she was fully gratified. In like 
manner, though the personal self be manifested as identical 
with the universal soul, its manifestation effects no com 
plete satisfaction so long as there is no recognition of those 
attributes ; but as soon as it is taught by a spiritual director 
to recognise in itself the perfections of MahesVara, his 
omniscience, omnipotence, and other attributes, it attains 
the whole pleroma of being. 

It is therefore said in the fourth section 
" As the gallant standing before the damsel is disdained 
as like all other men, so long as he is unrecognised, 
though he humble himself before her with all 
manner of importunities : In like manner the per 
sonal self of mankind, though it be the universal 
soul, in which there is no perfection unrealised, 
attains not its own glorious nature ; and therefore 
this recognition thereof must come into play." 
This system has been treated in detail by Abhinava- 
gupta and other teachers, but as we have in hand a sum 
mary exposition of systems, we cannot extend the discus 
sion of it any further lest our work become too prolix. 
This then may suffice. 1 A. E. G. 

[ l I have seen in Calcutta a short the son of Udayakara (cf. pp. 130, 
Comin. on the !Siva sutras by Utpala, 131 ). E. B. C. J 

( 137 ) 



OTHER Mahesvaras there are who, while they hold the 
identity of self with God, insist upon the tenet that the 
liberation in this life taught in all the systems depends 
upon the stability of the bodily frame, and therefore 
celebrate the virtues of mercury or quicksilver as a means 
of strengthening the system. Mercury is called pdrada, 
because it is a means of conveyance beyond the series of 
transmigratory states. Thus it has been said 

" It gives the farther shore of metempsychosis : it is 

called pdrada." 
And again in the Easarnava 

"It is styled pdrada because it is employed for the 
highest end by the best votaries. 

" Since this in sleep identical with me, goddess, arises 
from my members, and is the exudation of my 
body, it is called rasa." 

It may be urged that the literal interpretation of these 
words is incorrect, the liberation in this life being expli 
cable in another manner. This objection is not allowable, 
liberation being set out in the six systems as subsequent to 
the death of the body, and upon this there can be no 
reliance, and consequently no activity to attain to it free 
from misgivings. This is also laid down in the same 

1 Cf. Marco Polo s account of the the practices of the Siddhopdsakas 
Indian yogis in Colonel Yule s edit, in the Sankara-digvijaya, 49, to 
vol. ii. p. 3- Pdrada-pdna is one of obviate apamrityu, akdlamrityit, &c. 


" Liberation is declared in the six systems to follow the 

death of the body. 
" Such liberation is not cognised in perception like an 

emblic myrobalan fruit in the hand. 
" Therefore a man should preserve that body by means 

of mercury and of medicaments." 
Govinda-bhagavat also says 

"Holding that the enjoyments of wealth and of the 

body are not permanent, one should strive 
"After emancipation; but emancipation results from 
knowledge, knowledge from study, and study is 
only possible in a healthy body." 

The body, some one may say, is seen to be perishable, 
how can its permanency be effected ? Think not so, it is 
replied, for though the body, as a complexus of six sheaths 
or wrappers of the soul, is dissoluble, yet the body, as 
created by Hara and Gauri under the names of mercury 
and mica, may be perdurable. Thus it is said in the 

" They who, without quitting the body, have attained to 

a new body, the creation of Hara and Gauri, 
" They are to be lauded, perfected by mercury, at whose 

service is the aggregate of magic texts." 
The ascetic, therefore, who aspires to liberation in this 
life should first make to himself a glorified body. And 
inasmuch as mercury is produced by the creative conjunc 
tion of Hara and Gauri, and mica is produced from Gauri, 
mercury and mica are severally identified with Hara and 
Gauri in the verse 

"Mica is thy seed, and mercury is my seed; 

" The combination of the two, goddess, is destructive 

of death and poverty." 

This is very little to say about the matter. In the 
Base&varasiddhanta many among the gods, the Daityas, 
the Munis, and mankind, are declared to have attained to 
liberation in this life by acquiring a divine body through 
the efficacy of quicksilver. 


" Certain of the gods, Mahesa and others ; certain 
Daityas, Sukra and others ; 

"Certain Munis, the Balakhilyas and others; certain 
kings, Somesvara and others ; 

" Govinda-bhagavat, Govinda-nayaka, 

" Charvati, Kapila, Vyali, Kapali, Kandalayana, 

" These and many others proceed perfected, liberated 
while alive, 

"Having attained to a mercurial body, and therewith 

The meaning of this, as explicated by Paramesvara to 
Paramesvari, is as follows : 

"By the method of works is attained, supreme of 
goddesses, the preservation of the body ; 

11 And the method of works is said to be twofold, mer 
cury and air, 

" Mercury and air swooning carry off diseases, dead they 
restore to life, 

" Bound they give the power of flying about." 

The swooning state of mercury is thus described 

" They say quicksilver to be swooning when it is per 
ceived, as characterised thus 

" Of various colours, and free from excessive volatility. 

" A man should regard that quicksilver as dead, in which 
the following marks are seen 

" Wetness, thickness, brightness, heaviness, mobility." 

The bound condition is described in another place as 
follows : 

" The character of bound quicksilver is that it is 

" Continuous, fluent, luminous, pure, heavy, and that it 
parts asunder under friction." 

Some one may urge : If the creation of mercury by 
Hara and Gauri were proved, it might be allowed that the 
body could be made permanent; but how can that be 
proved ? The objection is not allowable, inasmuch as that 
can be proved by the eighteen modes of elaboration. Thus 
it is stated by the authorities 


"Eighteen modes of elaboration are to be carefully 

" In the first place, as pure in every process, for perfect 
ing the adepts." 
And these modes of elaboration are enumerated thus 

" Sweating, rubbing, swooning, fixing, dropping, coercion, 

" Kindling, going, falling into globules, pulverising, 

"Internal flux, external flux, burning, colouring, and 

"And eating it by parting and piercing it, are the 
eighteen modes of treating quicksilver." 

These treatments have been described at length by 
Govinda - bhagavat, Sarvajna - ramesvara and the other 
ancient authorities, and are here omitted to avoid pro 

The mercurial system is not to be looked upon as merely 
eulogistic of the metal, it being immediately, through the 
conservation of the body, a means to the highest end, 
liberation. Thus it is said in the Easarnava 

" Declare to me, god, that supremely efficacious 
destruction of the blood, that destruction of the body, 
imparted by thee, whereby it attained the power of flying 
about in the sky. Goddess (he replied), quicksilver is to 
be applied both to the blood and to the body. This makes 
the appearance of body and blood alike. A man should 
first try it upon the blood, and then apply it to the 

It will be asked : Why should we make this effort to 
acquire a celestial body, seeing that liberation is effected 
by the self-manifestation of the supreme principle, exist 
ence, intelligence, and beatitude ? We reply : This is no 
objection, such liberation being inaccessible unless we 
acquire a healthy body. Thus it is said in the Easah- 

" That intelligence and bliss set forth in all the systems 


in which a multitude of uncertainties are melted 
" Though it manifest itself, what can it effect for beings 

whose bodies are unglorified ? 

" He who is worn out with decrepitude, though he be 
free from cough, from asthma, and similar in 

" He is not qualified for meditation in whom the activi 
ties of the cognitive organs are obstructed. 
" A youth of sixteen addicted to the last degree to the 

enjoyment of sensual pleasures, 
" An old man in his dotage, how should either of these 

attain to emancipation ? " 

Some one will object : It is the nature of the personal 
soul to pass through a series of embodiments, and to be 
liberated is to be extricated from that series of embodi 
ments ; how, then, can these two mutually exclusive con 
ditions pertain to the same bodily tenement ? The objec 
tion is invalid, as unable to stand before the following 
dilemmatic argument : Is this extrication, as to the nature 
of which all the founders of institutes are at one, to be 
held as cognisable or as incognisable ? If it is incognisable, 
it is a pure chimera ; if it is cognisable, we cannot dispense 
with life, for that which is not alive cannot be cognisant of 
it. Thus it is said in the Kasasiddhanta 

l The liberation of the personal soul is declared in the 

mercurial system, subtile thinker. 
"In the tenets of other schools which repose on a 

diversity of argument, 
" Know that this knowledge and knowable is allowed 

in all sacred texts ; 

" One not living cannot know the knowable, and there 
fore there is and must be life." 

And this is not to be supposed to be unprecedented, 
for the adherents of the doctrine of Vishnu-svamin main 
tain the eternity of the body of Vishnu half-man and half- 
lion. Thus it is said in the Sakara-siddhi 


" I glorify the man-lion set forth by Vishnu- svamin, 
" Whose only body is existence, intelligence, and eternal 

and inconceivably perfect beatitude." 
If the objection be raised that the body of the man-lion, 
which appears as composite and as coloured, is incompatible 
with real existence, it may be replied : How can the body 
of the man-lion be otherwise than really existent, proved 
as it is by three kinds of proof: (i.) by the intuition of 
Sanaka and others ; (2.) by Vedic texts such as, A thousand 
heads has Purusha; and (3.) by Puranic texts such as, 
That wondrous child, lotus-eyed, four-armed, armed with 
the conch-shell, the club, and other weapons ? Eeal exist 
ence and other like predicates are affirmed also by Srikanta- 
misra, the devoted adherent of Vislmu-svamin. Let, then, 
those who aspire to the highest end of personal souls be 
assured that the eternity of the body which we are setting 
forth is by no means a mere innovation. It has thus 
been said 

" What higher beatitude is there than a body undecay- 

ing, immortal, 
" The repository of sciences, the root of merit, riches, 

pleasure, liberation ? " 

It is mercury alone that can make the body undecaying 
and immortal, as it is said 

" Only this supreme medicament can make the body mi- 
decaying and imperishable." 

Why describe the efficacy of this metal ? Its value is 
proved even by seeing it, and by touching it, as it is said 
in the Rasarnava 

"From seeing it, from touching it, from eating it, from 

merely remembering it, 
" From worshipping it, from tasting it, from imparting 

it, appear its six virtues. 
" Equal merit accrues from seeing mercury as accrues 

from seeing all the phallic emblems 
"On earth, those at Kedara, and all others whatso 


In another place we read 

" The adoration of the sacred quicksilver is more beatific 
than the worship of all the phallic emblems at 
Kasi and elsewhere, 
" Inasmuch as there is attained thereby enjoyment, 

health, exemption from decay, and immortality." 
The sin of disparaging mercury is also set out 

" The adept on hearing quicksilver heedlessly disparaged 

should recall quicksilver to mind. 
" He should at once shun the blasphemer, who is by his 

blasphemy for ever filled with sin." 
The attainment, then, of the highest end of the per 
sonal soul takes place by an intuition of the highest prin 
ciple by means of the practice of union (efoxjt?) after the 
acquisition of a divine body in the manner we have de 
scribed. Thereafter 

" The light of pure intelligence shines forth unto certain 

men of holy vision, 
" Which, seated between the two eyebrows, illumines 

the universe, like fire, or lightning, or the sun : 
" Perfect beatitude, unalloyed, absolute, the essence 

whereof is luminousness, undifferenced, 
"From which all troubles are fallen away, knowable, 

tranquil, self-recognised : 
" Fixing the internal organ upon that, seeing the whole 

universe manifested, made of pure intelligence, 
" The aspirant even in this life attains to the absolute, 

his bondage to works annulled." 

A Vedic text also declares : That is Rasa (mercury), 
having obtained this he becomes beatitude. 

Thus, then, it has been shown that mercury alone is the 
means of passing beyond the burden of transmigratory 
pains. And conformably we have a verse which sets 
forth the identity between mercury and the supreme self 
that mercury, which is the very self, preserve us 
dejection and from the terrors of metem 


"Which is naturally to be -applied again and again by 

those that aspire to liberation from the enveloping 

" Which perfected endures, which plays not again when 

the soul awakes, 
"Which, when it arises, pains no other soul, which 

shines forth by itself from itself." A. E. Gr. 



WHOSO wishes to escape the reality of pain, which is 
established by the consciousness of every soul through its 
being felt to be essentially contrary to every rational 
being, and wishes therefore to know the means of such 
escape, learns that the knowledge of the Supreme Being 
is the true means thereof, from the authority of such pas 
sages as these (vetdvatara Upan. vi. 20) _ 

" When men shall roll up the sky as a piece of leather, 

" Then shall there be an end of pain without the know 
ledge of Siva." 

Now the knowledge of the Supreme is to be gained by 
hearing (sravana), thought (manaTia), and reflection (bhd- 
vana), as it has been said 

" By scripture, by inference, and by the force of repeated 

" By these three methods producing knowledge, he gains 
the highest union (yoga)" 

Here thought depends on inference, and inference de 
pends on the knowledge of the vydpti (or universal pro 
position), and the knowledge of the vydpti follows the 
right understanding of the categories, hence the saint 
Kanada 2 establishes the six categories in his tenfold 

i ^Vaiseshikas are called Aulu- 1. 23), Akshapdda, Kanaka, Ulfika 

ky ah m Hemachandra s A bhidhdna- and Vatsa are called the sons of Siva 

Cfontdmani; m the Vayu-purana * He is here called by his synonym 

(quoted m Aufrecht s Catal. p. 53 b, Kanabhaksha. 



treatise, commencing with the words, " Now, therefore, we 
shall explain duty." 

In the first book, consisting of two daily lessons, he 
describes all the categories which are capable of intimate 
relation. In the first dhnika he defines those which pos 
sess "genus" (jdti), in the second "genus" (or "generality") 
itself and " particularity." In the similarly divided second 
book he discusses " substance," giving in the first dhnika 
the characteristics of the five elements, and in the second 
he establishes the existence of space and time. In the 
third book he defines the soul and the internal sense, the 
former in the first dhnika, the latter in the second. In 
the fourth book he discusses the body and its adjuncts, 
the latter in the first dhnika, and the former in the second. 
In the fifth book .he investigates action ; in the first dhnika 
he considers action as connected with the body, in the 
second as belonging to the mind. In the sixth book he 
examines merit and demerit as revealed in Sruti ; in the 
first dhnika he discusses the merit of giving, receiving 
gifts, &c., in the second the duties of the four periods of 
religious life. In the seventh book he discusses quality 
and intimate relation ; in the first dhnika he considers the 
qualities independent of thought, in the second those 
qualities which are related to it, and also intimate rela 
tion. In the eighth book he examines " indeterminate " 
and " determinate " perception, and means of proof. In 
the ninth book he discusses the characteristics of intellect. 
In the tenth book he establishes the different kinds of 
inference. 1 

The method of this system is said to be threefold, 
"enunciation," "definition," and "investigation." J " But," 
it may be objected, " ought we not to include division, 

1 It is singular that this is in- difference of the qualities of the 
accurate. The ninth book treats of soul, and the three causes, 
that perception which arises from * For this extract from the old 
supersensible contact, &c, and infer- IMshya of Vdtsyayana, see Cole- 
ence The tenth treats of the mutual brooke s Essays (new edition), vol. i. 

p. 285. 


and so make the method fourfold, not threefold ? " We 
demur to this, because " division " is really included in a 
particular kind of enunciation. Thus when we declare 
that substance, quality, action, generality, particularity, and 

intimate relation are the only six positive categories, 

this is an example of enunciation. If you ask " What is 
the reason for this definite order of the categories ? " we 
answer as follows : Since "substance" is the chief, as being 
the substratum of all the categories, we enounce this first; 
next "quality," since it resides in its generic character in 
all substances [though different substances have different 
qualities] ; then " action," as it agrees with " substance " 
and " quality " in possessing " generality ; " l then " gener 
ality," as residing in these three; then "particularity," 
inasmuch as it possesses "intimate relation;" 2 lastly, 
" intimate relation " itself ; such is the principle of arrange 

If you ask, " Why do you say that there are only six 
categories since non-existence is also one 1 " we answer : 
Because we wish to speak of the six as positive categories, 
i.e., as being the objects of conceptions which do not 
involve a negative idea. " Still," the objector may retort, 
"how do you establish this definite number only six ? 
for either horn of the alternative fails. For, we ask, is 
the thing to be thus excluded already thoroughly ascer 
tained or not ? If it is thoroughly ascertained, why do you 
exclude it ? and still more so, if it is not thoroughly 
ascertained ? What sensible man, pray, spends his strength 
in denying that a mouse has horns ? Thus your definite 
number only six fails as being inapplicable." This, how 
ever, we cannot admit ; if darkness, &c., are allowed to 
form certainly a seventh category (as " non-existence ), 
we thus (by our definite number) deny it to be one of the 
six positive categories, and if others attempt to include 

1 Cf. Bhdshd-parichclJieda, sloka by " intimate relation " in the eter- 
!4- nal atoms, &c. 

2 "Particularity " (visesha) resides 


" capacity," " number," &c., which we allow to be certainly 
positive existences, we thus deny that they make a seventh 
category. But enough of this long discussion. 

Substantiality, &c. (dravyatvddi), i.e., the genera of sub 
stance, quality, and action, are the definition of the triad 
substance, quality, and action respectively. The genus of 
substance (dravyatva) is that which, while it alike exists 
with intimate relation in the (eternal) sky and the (tran 
sitory) lotus, is itself eternal, 1 and does not exist with 
intimate relation in smell. 2 

The genus of quality (gunatva) is that which is imme 
diately subordinate to the genus existence, and exists with 
intimate relation in whatever is not an intimate or mediate 
cause. 3 The genus of action (Jcarmatva) is that which is 
immediately subordinate to the genus existence, and is 
not found with intimate relation in anything eternal. 4 
Generality (or genus, sdmdnya) is that which is found in 
many things with intimate relation, and can never be the 
counter-entity to emergent non-existence. 5 Particularity 6 
(visesha) exists with intimate relation, but it is destitute 

1 This clause is added, as other- the MS. in the Calcutta Sanskrit 
wise the definition would apply to College Library. 

"duality" and "conjunction." 5 I.e., it can never be destroyed. 

2 This is added, as otherwise the Indestructibility, however, is found 
definition would apply to "exist- in time, space, &c. ; to exclude these, 
ence " (sattd), which is the summum therefore, the former clause of the 
genus, to which substance, quality, definition is added. 

and action are immediately sub- 6 " Particularity " (whence the 

ordinate. name Vaiseshika) is not "individu- 

3 Existence (sattd) is the genus of ality, as of this particular flash of 
dravya, guna, and kriyd. Dravya lightning," but it is the individu- 
alone can be the intimate cause of ality either of those eternal sub- 
anything ; and all actions are the stances which, being single, have no 
mediate (or non-intimate) cause of genus, as ether, time, and space ; 
conjunction and disjunction. Some or of the different atomic minds ; or 
qualities (as samyoga, rupa, &c.) of the atoms of the four remaining 
may be mediate causes, but this is substances, earth, water, fire, and 
accidental and does not belong to air, these atoms being supposed to be 
the essence of guna. as many gunas the ne plus ultra, and as they have 
can never be mediate causes. no parts, they are what they are by 

4 As all karmas are transitory, their own indivisible nature. Ballan- 
Jcarmatva is only found in the anitya, t} r ne translated visesha as "ultimate 
I correct in p. 105, line 20, nitijd- difference." I am not sure whether 
s&mavetatva ; this is the reading of the individual soul hi 


of generality, which stops mutual non-existence. 1 Intimate 
relation (samavdya) is that connection which itself has 
not intimate relation. 2 Such are the definitions of the 
six categories. 

Substance is ninefold, earth, water, fire, air, ether, time, 
space, soul, and mind. The genera of earth, &c. (prithi- 
vitva), are the definitions of the first four. The genus of earth 
is that generality which is immediately subordinate to 
substance, and resides in the same subject with colour 
produced by baking. 3 

The genus of water is that generality which is found 
with intimate relation in water, being also found in intimate 
relation in river and sea. The genus of fire is that gener 
ality which is found with intimate relation in fire, being 
also found with intimate relation in the moon and gold. 
The genus of air is that which is immediately subordinate 
to substance, and is found with intimate relation in the 
organ of the skin. 4 

As ether, space, and time, from their being single, can 
not be subordinate genera, their several names stand 
respectively for their technical appellations. Ether is the 
abode of particularity, and is found in the same subject 
with the non-eternal (janya) special quality which is not 
produced by contact. 5 

Time is that which, being a pervading substance, is the 
abode of the mediate cause 6 of that idea of remoteness 

1 Mutual non-existence (anyonyd- 4 The organ of touch is an aerial 
IMva) exists between two notions integument. Colebrooke. 

which have no property in common, 5 Sound is twofold, "produced 

as a "pot is not cloth;" but the from contact," as the first sound, and 

genus is the same in two pots, both "produced from sound," as the 

alike being pots. second. Janya is added to exclude 

2 " SamavdyasambanddbJidvdt sa- God s knowledge, while samyogd- 
mavdyo na jdtih," Siddh. Mukt. janya excludes the soul s, which is 
(Sarnyoya being a guna has gunatva produced by contact, as of the soul 
existing in it with intimate rela- and mind, mind and the senses, &c. 
tion). 6 The mediate cause itself is the 

3 The feel or touch of earth is said conjunction of time with some body, 
to be "neither hot nor cold, and its &c., existing in time, this latter is 
colour, taste, smell, and touch are the intimate cause, while the know- 
changed by union with fire" (Bhd- ledge of the revolutions of the sun 
shaparichchheda, si. 103, 104). is the instrumental cause. In p. 

106, line 12, read adliikaranam. 


(paratva) which is not found with intimate relation in 
space ; l while space is that pervading substance which pos 
sesses no special qualities and yet is not time. 2 The general 
terms dtmatva and manastva are the respective definitions 
of soul (dtmaii) and mind (manas). The general idea of soul 
is that which is subordinate to substance, being also found 
withintimate relation in that which is without form 3 amtirt- 
ta). The general idea of mind is that which is subordinate 
to substance, being also found existing with intimate rela 
tion in an atom, but [unlike other atoms] not the intimate 
cause of any substance. There are twenty-four qualities ; 
seventeen are mentioned directly in Kanada s Sutras (i. 1,6), 
" colour, taste, smell, touch, number, quantity, severalty, 
conjunction, disjunction, remoteness, proximity, intelli 
gence, pleasure, pain, desire, aversion, and effort;" and, 
besides these, seven others are understood in the word 
"and," viz., gravity, fluidity, viscidity, faculty, merit, 
demerit, and sound. Their respective genera (rupatva, 
&c.) are their several definitions. The class or genus of 
" colour " is that which is subordinate to quality and exists 
with intimate relation in blue. In the same way may be 
formed the definitions of the rest. 

"Action" is fivefold, according to the distinction of 
throwing upwards, throwing downwards, contracting, ex 
panding, and going: revolution, evacuating, &c., being 
included under " going." The genus of throwing upwards, 
&c., will be their respective definitions. The genus of 
throwing upwards is a subordinate genus to action; it 
exists with intimate relation, and is to be known as 
the mediate cause of conjunction with a higher place. In 
the same manner are to be made the definitions of throw 
ing downwards, &c. Generality (or genus) is twofold, 
extensive and non-extensive ; existence is extensive as 
found with intimate connection in substance and quality, 

1 Paratva being of two kinds, ever, is not pervading but atomic. 
daisika and Jcdlika. 3 The three other paddrthas, beside 

2 Time, space, and mind have soul, which are amtirtta, time, ether, 
no special qualities ; the last, how- and space, are not genera. 


or in quality and action ; substance, &c., are non-extensive. 
The definition of generality has been given before. Par 
ticularity and intimate relation cannot be divided, in 
the former case in consequence of the infinite number of 
separate particularities, in the latter from intimate relation 
being but one ; their definitions have been given before. 

There is a popular proverb 

" Duality, change produced by baking, and disjunction 
produced by disjunction, he whose mind vacillates not in 
these three is the true Vaiseshika ; " and therefore we will 
now show the manner of the production of duality, &c. 

There is here first the contact of the organ of sense 
with the object ; thence there arises the knowledge of the 
genus unity ; then the distinguishing perception apekshd- 
luddhi [by which we apprehend "this is one," "this is 
one," &c.] ; then the production of duality, dvitm (in the 
object); 1 then the knowledge of the abstract genus of 
duality (dmtvatva) ; then the knowledge of the quality 
duality as it exists in the two things ; then imagination 2 

But it may here be asked what is the proof of duality, 
&c., being thus produced from apekshdbuddhi ? The great 
doctor (Udayana) maintained that apeJcshdbuddhi must be 
the producer of duality, &c., because duality is never 
found separated from it, while, at the same time, we 
cannot hold apekshdbuddhi as the cause only of its being 
known [and therefore it follows that it must be the cause 
of its being produced 4 ], just as contact is with regard to 
sound. We, however, maintain the same opinion by a 

1 All numbers, from duality up- material previously supplied to it by 
wards, are artificial, i.e., they are the senses and the internal organ or 
made by our minds; unity alone mind. (Of. the tables in p. 153.) 
exists in things themselves each 3 Here and elsewhere I omit the 
being one ; and they only become metrical summary of the original, as 
two, &c., by our choosing to regard it adds nothing new to the previous 
them so, and thus joining them in prose. 

thought. 4 Every cause must be either 

2 SamsJcdra is here the idea con- jMpdka or jawka ; apelcshdfiuddhi, 
ceived by the mind created, in not being the former, must be the 
fact, by its own energies out of the latter. 


different argument ; duality, &c., cannot be held to be made 
known (jndpya) by that non-eternal apprehension whose 
object is two or more individual unities (i.e., apekshdluddhi), 
because these are qualities which reside in a plurality of 
subjects [and not in any one individual 1 ] just as "seve- 
ralty " does [and, therefore, as apeksMbuddhi is not their 
jndpaka, it must be their janaTca]. 

Next we will describe the order of the successive destruc 
tions. From apeksMbuddhi arises, simultaneously with the 
production of duality (dvitva), the destruction of the know 
ledge of the genus of unity; next from the knowledge of 
the genus of duality (dvitvatw) arises, simultaneously with 
the knowledge of the quality duality, the destruction of 
apeksMbuddhi; next from the destruction of apeksMbuddhi 
arises, simultaneously with the knowledge of the two sub 
stances, the destruction of the duality; next from the 
knowledge of the two substances arises, simultaneously 
with the production of imagination (saniskdra), the destruc 
tion of the knowledge of the quality; and next from 
imagination arises the destruction of the knowledge of the 

The evidence for the destruction of one kind of know 
ledge by another, and for the destruction of another know 
ledge by imagination, is to be found in the following 
argument; these knowledges themselves which are the 
subjects of the discussion are successively destroyed by 
the rise of others produced from them, because knowledge, 
like sound, is a special quality of an all-pervading sub 
stance, and of momentary duration. 2 I may briefly add, 
that when you have the knowledge of the genus of unity 
simultaneously with an action in one of the two things 
themselves, producing that separation which is the opposite 

1 Apekshdbuddhi apprehends "this pervading substance, but the in- 

is one," "this is one," &c. ; but dividual portions of each have differ- 

duahty, for instance, does not reside ent special qualities ; hence one man 

m either of these, but in loth to- knows what another is ignorant of, 

ST 1 ^/ ^ " t and one portion of ether has sound 

t The Vaiseshikas held that the when another portion has not. Dr. 
jivatman and space are each an all- Roer, in his version of the Bhasha" 


to the conjunction that produced the whole, in that 
case you have the subsequent destruction of duality pro 
duced by the destruction of its abiding-place (the two 
things) ; but where you have this separate action taking 
place simultaneously with the rise of apekshdbuddhi, there 
you have the destruction of duality produced by the 
united influence of both. 1 

Apekshabuddhi is to be considered as that operation of 
the mind which is the counter-entity to that emergent 
non-existence (i.e., destruction) which itself causes a sub 
sequent destruction. 2 

Parichchheda, has mistranslated an 
important Sutra which bears on this 
point. It is said in Sutra 26 


avyapyavrittih kshaniko visesha- 

guna ishyate, 

which does not mean "the special 
qualities of ether and soul are limi 
tation to space and momentary dura 
tion," but "the special qualities of 

1. Ekatva-jiiana . . 

2. Apekshabuddhi . 

3. Dvitvotpatti and ek 


4. Dvitvatvajiiana . 

5. Dvitvaguna - buddhi 

and apekshabud- 
dhindsa .... 

6. Dvitva - ndsa and 

dravya-buddhi . . 

The second and third columns 
represent what takes place when, in 
the course of the six steps of ekatva- 
jndna, &c., one of the two parts 
is itself divided either at the first 
or the second moment. In the first 
case, the dvitva of the whole Is de 
stroyed in the fifth moment, and 
therefore its only cause is its imme 
diately preceding dvitvddhdra-ndsa, 
or, as Mdclhava calls it, dsrayaniv- 
ritti. In the second case, the ndsa 
arrives at the same moment simul 
taneously by both columns (j) and 
(3), and hence it may be ascribed to 

ether and soul (i.e., sound, know 
ledge, &c.) are limited to different 
portions and of momentary dura 

1 The author here mentions two 
other causes of the destruction of 
dvitva besides that already given 
in p. 152, 1. 14 (apekshdluddhi-ndta}, 
viz., dsrayandsa, and the united action 
of loth : 

Avayava-kriya . . . 
Avayava-vibhaga . . 
Avayava - samyoga- 
Dvitvadharasya (z .e., 
avayavinah) ndsah 
Dvitva - nasa (i.e., of 
avayavin). . . .. 




Adhdra-ndsa (of i 


the united action of two causes, 
apekshdbuddhi-ndsa and ddkdra- ndta. 
Any.kriyd which arose in one of the 
parts after the second moment 
would be unimportant, as the ndsa 
of the dvitva of the whole would t take 
place by the original sequence in 
column (i) in the sixth moment; 
and in this way it would be too late 
to affect that result. 

2 I.e., from the destruction of 
apekshdbuddhi follows the destruc 
tion of dvitva ; but the other destruc 
tions previously described were fol 
lowed by some production, thus 


Next we will inquire in how many moments, commenc 
ing with the destruction of the compound of two atoms (the 
dvyanuka), another compound of two atoms is produced, 
having colour, &c. In the course of this investigation the 
mode of production will be explained. First, the com 
pound of two atoms is gradually destroyed by the series 
of steps commencing with the contact of fire ; l secondly, 
from the conjunction of fire arises the destruction of the 
qualities black, &c., in the single atom; thirdly, from 
another conjunction of fire arises the production of red, 
&c., in the atom ; fourthly, from conjunction with a soul 
possessing merit arises an action 2 in the atom for the 
production of a substance ; fifthly, by that action is pro 
duced a separation of that atom from its former place; 
sixthly, there is produced thereby the destruction of its 
conjunction with that former place ; seventhly, is produced 
the conjunction with another atom ; eighthly, from these 
two atoms arises the compound of two atoms; ninthly, 
from the qualities, &c., of the causes (i.e., the atoms) are 
produced colour, &c., the qualities of the effect (i.e., the 
dvyaiiu ka). Such is the order of the series of nine mo 
ments. The other two series, 3 that of the ten and that of 
the eleven moments, are omitted for fear of prolixity. 
Such is the mode of production, if we hold (with the 
Vaiseshikas) that the baking process takes place in the 

the knowledge of dvitvatra arose length in the Siddhdnta Muktavali, 

from the destruction of ekatvajndna, pp. 104, 105. In the first series we 

&c. (cf. Siddd. Mukt., p. 107). I have I . the destruction of the dvya- 

may remind the reader that in Hindu nuka and simultaneously a disjunc- 

logic the counter-entity to the non- tion from the old place produced by 

existence of a thing is the thing itself . the disjunction (of the parts); 2. 

1 From the conjunction of fire is the destruction of the black colour 
produced an action in the atoms of in the dvyanuka, and the simul- 
the jar ; thence a separation of one taneous destruction of the conjunc- 
atom from another , thence a de- tion of the dvyanuka with that place ; 
struction of the conjunction of atoms 3. the production of the red colour 
which made the black (or unbaked) in the atoms, and the simultaneous 
jar ; thence the destruction of the conjunction with another place ; 4. 
compound of two atoms. the cessation of the action in the 

2 I.e., a kind of initiative ten- atom produced by the original con- 
dency. junction of fire. The remaining 

3 These are explained at full 5-10 agree with the 4-9 above. 


atoms of the jar. 1 The Naiyayikas, however, maintain 
that the baking process takes place in the jar. 

" Disjunction produced by disjunction " is twofold, 
that produced by the disjunction of the intimate [or 
material] causes only, and that produced by the disjunction 
of the intimate cause and the non-cause [i.e., the place]. 
We will first describe the former kind. 

It is a fixed rule that when the action of breaking arises 
in the [material] cause which is inseparably connected 
with the effect [i.e., in one of the two halves of the pot], 
and produces a disjunction from the other half, there is 
not produced at that time a disjunction from the place or 
point of space occupied by the pot ; and, again, when there 
is a disjunction from that point of space occupied by the 
pot, the disjunction from the other half is not contem 
porary with it, but has already taken place. For just as 
we never see smoke without its cause, fire, so we never see 
that effect of the breaking in the pot which we call the 
disjunction from the point of space, 2 without there having 
previously been the origination of that disjunction of the 
halves which stops the conjunction whereby the pot was 
brought into being. Therefore the action of breaking in 
the parts produces the disjunction of one part from another, 
but not the disjunction from the point of space ; next, this 
disjunction of one part from another produces the destruc 
tion of that conjunction which had brought the pot into 
existence ; and thence arises the destruction of the pot, 
according to the principle, cessante causa cessat efectus. 
The pot being thus destroyed, that disjunction, which 

1 The Vaiseshikas hold that when followers of the Nyaya maintain that 

a jar is baked, the old black jar is the fire penetrates into the different 

destroyed, its several compounds of compounds of two or more atoms, 

two atoms, &c., being destroyed ; and, without any destruction of the 

the action of the fire then produces old jar, produces its effects on these 

the red colour in the separate atoms, compounds, and thereby changes not 

and, joining these into new com- the jar but its colour, &c., it is still 

pounds, eventually produces a new the same jar, only it is red, not 

red jar. The exceeding rapidity of black. 

the steps prevents the eye s detect- 2 In p. 109, line 14, I read gaga- 
ing the change of the jars. The navibhdgakartritrasya. 


resides in both the halves (which are the material or 
intimate causes of the pot) during the time that is marked 
by the destruction of the pot or perhaps having reference 
only to one independent half, initiates, in the case of 
that half where the breaking began, a disjunction from 
the point of space which had been connected with the 
pot ; but not in the case of the other half, as there is no 
cause to produce it. 1 

But the second kind is as follows : As action which 
arises in the hand, and causes a disjunction from that 
with which it was in contact, initiates a disjunction 2 from 
the points of space in which the original conjunction took 
place ; and this is " the disjunction of the intimate cause 
and the non-cause." When the action in the hand produces 
an effect in relation to any points of space, it initiates also 
in the same direction a disjunction of the intimate effect 
and the non-effect ; thus the disjunction of the body [the 
intimate effect] and the points of space arises from the dis 
junction of the hand and the points of space [the hand being 
an intimate or material cause of the body, but the points of 
space being not a cause]. This second disjunction is not 
produced by the action of the body, because the body is 
supposed to be at the time inactive ; nor is it produced by 
the action of the hand, because it is impossible that an 
action residing in some other place [as the hand] should 
produce the effect of disjunction [in the body]. Therefore 
we conclude by exhaustion that we must accept the view 
that it is the disjunction of the intimate cause and the 

1 The Siddhanta Muktclvali, p, 1 1 2, conjunction with that old place ; 7. 

describes the series of steps : I. An the conjunction with the new place ; 

action, as of breaking, in one of the 8. the cessation of the original im- 

halves ; 2. the disjunction of the pulse of fracture. Here the second 

two halves ; 3. the destruction of disjunction (viz., of the half of the 

the conjunction which originally pot and the place) is produced by 

produced the pot; 4. the destruc- the previous disjunction of the halves, 

tion of the pot ; 5. by the disjunction the intimate causes of the pot. 
of the two halves is produced a dis- 2 The original has a plural vi- 

junction of the severed half from the Nuigdn, i.e. , disjunctions from the 

old place ; 6. the destruction of the several points. 


non-cause 1 which causes the second disjunction of tKe 
body and the points of space. 

But an opponent may here object that " what you for 
merly stated (p. 147) as to existence being denied of dark 
ness, &c., is surely unreasonable ; for, in fact, there are no 
less than four different opinions maintained on this point, 
thus (a.) the Bhatta Mimamsakas and the Vedantins hold 
that darkness is a substance ; (&.) Sridhara Acharya 2 holds 
that the colour of dark blue is imposed [and thus darkness 
will be a quality] ; (e.) some of the Prabhakara Mimamsakas 
hold that it is the absence of the cognition of light ; (d.) 
the Naiyayikas, &c., hold that it is the absence of light." 
In reply, we assert that as for the first alleged opinion (a.) 
it is quite out of the question, as it is consistent with 
neither of the two possible alternatives ; for if darkness 
is a substance, it must either be one of the nine well- 
known substances, earth, &c., 3 or some different one. But 
it cannot be any one of the nine, since, under whichever 
one you would place it, all the qualities of that substance 
should certainly be found in it ; nor can you, on the other 
hand, assert that it is some substance different from these 
nine, since, being in itself destitute of qualities, it cannot 
properly be a substance at all [the very definition of sub 
stance being " that which is the substratum of qualities "], 
and therefore, of course, it cannot be a different substance 
from the nine. But you may ask, " How can you say that 
darkness is destitute of qualities, when it is perceived as 
possessed of the dark blue of the tamala blossom ? " We 
reply, that this is merely an error, as when men say that 
the [colourless] sky is blue. But enough of this onslaught 
on ancient sages. 4 (5.) Hence it follows that darkness can 
not have its colour imposed upon it, since you cannot have 
an imposition of colour without supposing some substratum 

1 I.e., the disjunction of the hand 4 I am not sure that it would not 
and the points of space. be better to read viddkarevidkayd, 

2 The author of a commentary on rewounding the wounded, instead of 
the Bhagavad Gita. vriddhaviiadhayd. 

3 For dravyddi read prithivyddi. 


to receive it ; l and again, we cannot conceive the eye as 
capable of imposing a colour when deprived of the con 
current cause, the external light. JSTor can we accept that 
it is an impression independent of the eye [i.e., produced 
by the internal sense, mind], because the concurrence of 
the eye is not a superfluous but an indispensable condi 
tion to its being produced. Nor can you maintain that 
" absence or non-existence (abhdva 2 ) is incapable of being 
expressed by affirmative tense affixes [and, therefore, as we 
do use such phrases as tenebrce oriuntur, darkness cannot 
be a mere non-existence "] ; because your assertion is too 
broad, as it would include such cases of non-existence as a 
mundane collapse, destruction, inattention, 3 &c. [and yet 
we all know that men do speak of any of these things as 
past, present, or future, and yet all are cases of abhdva]. 
(c.) Hence darkness cannot be the absence of the cognition of 
light, since, by the well-known rule that that organ which 
perceives a certain object can also perceive its absence, it 
would follow that darkness would be perceived by the 
mind [since it is the mind which perceives cognitions]. 4 
Hence we conclude that the fourth or remaining opinion 
must be the true one, viz., that darkness is only the 
absence of light. And it need not be objected that it is 
very difficult to account for the attribution to non-exist 
ence of the qualities of existence, for we all see that the 
quality happiness is attributed to the absence of pain, and 
the idea of separation is connected with the absence of 
conjunction. And you need not assert that " this absence 
of light must be the object of a cognition produced by the 
eye in dependence on light, since it is the absence of an 
object possessing colour, 5 as we see in the case of a jar s 

1 Unless you see the rope you can- dhaka-Tcriyd. It has that meaning 
not mistake it for a serpent. in Kavyaprakasa, V. (p. 114, 1. i). 

2 In p. no, last line, read bhdve. 4 The mind perceives dloka-jndna, 

3 Read in p. no, last Jine, anava- therefore it would perceive its ab- 
dhdnddishu. Vidhipralyaya properly sence, i.e., darkness, but this last is 
means an imperative or potential perceived by the eye. 

affix implying " command ; " but the 5 I.e., light possesses colour, and we 
pandit takes vidhi here as bhdvabo- cannot see a jar s absence in the dark. 


absence," because by the very rule on which you rely, viz., 
that that on which the eye depends to perceive an object, 
it must also depend on to perceive that object s absence, 
it follows that as there is no dependence of the eye on 
light to perceive light, it need not depend thereon to per 
ceive this light s absence. Nor need our opponent retort 
that " the cognition of darkness [as the absence of light] 
necessitates the cognition of the place where the absence 
resides [and this will require light]," as such an assertion 
is quite untenable, for we cannot admit that in order to 
have a conception of absence it is necessary to have a 
conception of the place where the absence resides, else 
we could not have the perception of the cessation of sound, 
as is implied in such an expression as " the tumult has 
ceased." 1 Hence, having all these difficulties in his mind, 
the venerable Kanada uttered his aphorism [as an ipse 
dixit to settle the question] : " Dravya-guna-karma-nisli- 
patti-vaidharmydd alhdvas tamas" (Vais. Sut. v. 2, 19), 
" Darkness is really non-existence, since it is dissimilar to 
the production of substances, qualities, or actions." The 
same thing has been also established by the argument that 
darkness is perceived by the eye 2 [without light, whereas 
all substances, if perceptible at all, require the presence 
of light as well as of the eye to be visible]. 

Non-existence (abhdva) is considered to be the seventh 
category, as established by negative proofs. It may be 
concisely denned as that which, itself not having intimate 
relation, is not intimate relation ; 3 and this is twofold, 
" relative non-existence " 4 and " reciprocal non-existence." 

1 Sound resides in the impercep- eva vd tamah sydt, vdhydlokapragra- 

tible ether, and cessation is the Jtam antarena chakshushd no, grill 

dhvamsdbhdva, or "emergent non- yeta." 

existence." 3 Intimate relation has also no 

2 The reading pratyayavedyatvena intimate relation. 

seems supported by p. no, last line, 4 "Relative non-existence" (sam- 

but it is difficult to trace the argu- sargdbhdra) is the negation of a 

ment ; I have, therefore, ventured relation ; thus " the jar is not in the 

hesitatingly to read pratyaksliave- house" is "absolute non-existence," 

dyatrena, and would refer to the " it was not in the house " is " ante- 

commeiitary (Vais. Sut. p. 250), cedent," and "it will not be in the 

" yadi hi nila-rupavan nilam rupam house "is "emergent," non-existence. 


The former is again divided into "antecedent," "emer 
gent/ 5 and " absolute." " Antecedent " is that non-exist 
ence which, though without any beginning, is not ever 
lasting; "emergent" is that which, though having a 
beginning, is everlasting ; " absolute " is that non-existence 
which abides in its own counter-entity ; l " reciprocal non- 
existence" is that which, being different from "absolute," 
has yet no defined limit [i.e., no terminus ad quern nor ter 
minus a quo, as " antecedent " and " emergent " have]. 

If you raise the objection that " reciprocal non-exist 
ence is really the same as absolute non-existence, " we 
reply that this is indeed to lose one s way in the king s 
highroad ; for " reciprocal non-existence " is that negation 
whose opposite is held to be identity, as "ajar is not cloth;" 
but " absolute non-existence " is that negation whose 
opposite is connection, as " there is no colour in the air." 2 
Nor need you here raise the objection that " abhdva can 
never be a means of producing any good to man," for we 
maintain that it is his summum bonum, in the form of 
final beatitude, which is only another term for the absolute 
abolition of all pain [and therefore comes under the cate 
gory of abhdva]. E. B. C. 

1 I.e., the absolute absence of the jdti gliatatva which resides in the 

jar is found in the jar, as, of course, jar. 

the jar does not reside in the jar, 2 The opposite is "there is colour 

but in the spot of ground, it is the in the air." 




THE principle that final bliss, i.e., the absolute abolition of 
pain, arises from the knowledge of the truth [though in a 
certain sense universally accepted], is established in a 
special sense as a particular tenet l of the Nyaya school, 
as is declared by the author of the aphorisms in the words 
" proof, that which is to be proved, &c., from knowledge 
of the truth as to these things there is the attainment of 
final bliss." This is the first aphorism of the Nydya 
Sastra. Now the Nyaya Sastra consists of five books, 
and each book contains two " daily portions." In the 
first daily portion of the first book the venerable Gotama 
discusses the definitions of nine categories, beginning with 
" proof," and in the second those of the remaining seven, 
beginning with "discussion" (vdda). In the first daily 
portion of the second book he examines " doubt," discusses 
the four kinds of "proof," and refutes the suggested 
objections to their being instruments of right knowledge; 
and in the second he shows that " presumption," &c., are 
really included in the four kinds of " proof " already given 
[and therefore need not be added by the Mimamsakas as 
separate ones]. In the first daily portion of the third 
book he examines the soul, the body, the senses, and their 
objects; in the second, "understanding" (buddhi), and 
" mind " (manas). In the first daily portion of the fourth 
book he examines "volition" (pmvritti), the "faults," 

1 Cf. Nydya Sutras, i. 29. 


" transmigration," " fruit " [of actions], " pain," and " final 
liberation ; " in the second he investigates the truth 1 as 
to the causes of the " faults," and also " wholes " and 
" parts." In the first daily portion of the fifth book he 
discusses the various kinds of futility (jdti), and in the 
second the various kinds of " occasion for rebuke " (nigra- 
hasthdna, or " unfitness to be argued with "). 

In accordance with the principle that " to know the 
thing to be measured you must first know the measure," 
"proof" (pramdna) is first enunciated, and as this must 
be done by defining it, we have first a definition of " proof." 
"Proof" is that which is always accompanied by right 
knowledge, and is at the same time not disjoined from 
the proper instruments [as the eye, &c.], and from the 
site of knowledge [i.e., the soul] ; 2 and this definition thus 
includes the peculiar tenet of the Nyaya School that God 
is a source of right knowledge, 3 as the author of the 
aphorisms has expressly declared (ii. 68), " and the fact 
of the Yeda s being a cause of right knowledge, like spells 
and the medical science, follows from the fact that the fit 
one who gave the Veda was a source of right knowledge." 
And thus too hath the universally renowned teacher 
Udayana, who saw to the farthest shore of the ocean of 
logic, declared in the fourth chapter of the Kusumanjali : 

" Eight knowledge is accurate comprehension, and right 
knowing is the possession thereof; authoritativeness is, 
according to Gotama s school, the being separated from all 
absence thereof. 

"He in whose intuitive unerring perception, insepar 
ably united to Him and dependent on no foreign inlets, 
the succession of all the various existing objects is con 
tained, all the chaff of our suspicion being swept away 

1 In p. 112, line 16, of the Cal- (visliaya), as these are, of course, 
cutta edition, I read doshanimitta- connected with right knowledge. 
tattva for doshanimittakatva (compare 3 Tsvara is a cause of right know- 
Nya"ya Sut. iv. 68). ledge (pramdna) according to the 

2 Without this last clause the definition, because he is pramdyd 
definition might include the objects dsrayah. 


by the removal of all possible faults as caused by the 
slightest want of observation in Him, He, Siva, is my 
authority; what have I to do with others, darkened as 
their authority must ever be with rising doubts ? " 

" Proof " is fourfold, as being divided into perception, 
inference, analogy, and testimony. The " thing to be 
proved" [or the "object of right notion"] is of twelve 
kinds, viz., soul, body, the senses, their objects, under 
standing, mind, volition, faults, transmigrations, fruit, pain, 
and final liberation. "Doubt" is a knowledge whose 
nature is uncertainty; and this is threefold, as being 
caused by the object s possessing only qualities which are 
common to other things also, and therefore not distinctive, 
or by its possessing only irrelevant qualities of its own, 
which do not help us in determining the particular point 
in question, 1 or by conflicting testimony. The thing which 
one proposes to one s self before proceeding to act, is " a 
motive " (prayojana) ; this is twofold, i.e., visible and 
invisible. " An example " is a fact brought forward as a 
ground for establishing a general principle, and it may 
be either affirmative or negative. 2 A " tenet " (siddhdnta) 
is something which is accepted as being authoritatively 
settled as true ; it is of four kinds, as being " common to 
all the schools," "peculiar to one school," "a pregnant 
assumption " [leading, if conceded, to a further conclusion], 
and "an implied dogma" (i. 26-31). The "member" (of 
a demonstration) is a part of the sentence containing an 
inference for the sake of another ; and these are five, the 
proposition, the reason, the example, the application, and 
the conclusion (i. 32-38). "Confutation" (tarka, i. 39) is 
the showing that the admission of a false minor necessi 
tates the admission of a false major 3 (cf. Siit. i. 39, and 

1 On this compare Siddhanta- the smoke, is the confutation of there 
Muktavali, p. 115. being no fire in the hill" (Ballan- 

2 On these compare my note to tyne). Or, in other words, "the 
Colebrooke s Essays, vol. i. p. 315. mountain must have the absence-of- 

3 " Our coming to the conclusion smoke (vydpaka) if it has the ab- 
that there can be no smoke in the sence-of-fire (the false vydpya "). 
hill if there be no fire, while we see 


iv. 3) ; and this is of eleven kinds, as vydghata, dtmdsraya, 
itaretardsraya, &c. 

" Ascertainment " (nirnaya, i. 40) is right knowledge or 
a perception of the real state of the case. It is of four 
kinds as produced by perception, inference, analogy, or 
testimony. "Discussion" (vdda) is a particular kind of 
conversation, having as its end the ascertainment of truth 
(i. 41). "Wrangling" (jalpa) is the talk of a man only 
wishing for victory, who is ready to employ arguments 
for either side of the question (i. 42). "Cavilling" (vi- 
tandd) is the talk of a man who does not attempt to 
establish his own side of the question (i. 43). " Dialogue " 
(kathd) is the taking of two opposite sides by two dis 
putants. A "fallacy" is an inconclusive reason which is 
supposed to prove something, and this may be of five 
kinds, the "erratic," the "contradictory," the "uncertain," 
the " unproved," and the " precluded " or " mistimed " 
(Sut. i. 44-49). "Unfairness" (chhala) is the bringing 
forward a contrary argument by using a term wilfully in 
an ambiguous sense ; this is of three kinds, as there may 
be fraud in respjsct of a term, the meaning, or a meta 
phorical phrase (i. 50-54). "Futility" (jdti) is a self- 
destructive argument (i. 58). This is of twenty-four kinds 
(as described in the fifth book of the Nyaya aphorisms 
(1-38). "Occasion for rebuke" is where the disputant 
loses his cause [by stupidity], and this is of twenty-two 
kinds (as described in the fifth book of the aphorisms, 
44-67). We do not insert here all the minute sub-divi 
sions through fear of being too prolix, they are fully 
explained in the aphorisms. 

But here an objector may say, " If these sixteen topics, 
proof, &c., are all thus fully discussed, how is it that it has 
received the name of the Nyaya Sastra, [as reasoning, i.e., 
Nydya,? logic, properly forms only a small part of the topics 
which it treats of ? "] We allow the force of the objection; 
still as names are proverbially said to be given for some 
special reason, we maintain that the name Nyaya was 


rightly applied to Gotama s system, since " reasoning," or 
inference for the sake of another, is justly held to be a 
predominant feature from its usefulness in all kinds of 
knowledge, and from its being a necessary means for every 
kind of pursuit. So it has been said by Sarvajna, " This 
is the pre-eminent science of ISTyaya from its establishing 
our doctrines against opponents, and from its producing 
action ; " l and by Pakshila Swamin, " This is the science 
of reasoning (tinviksJiiki) divided into the different cate 
gories, proof, &c. ; the lamp of all sciences, the means 
for aiding all actions, the ultimate appeal of all religious 
duties, well proved in the declarations of science." 2 

But here an objector may say, " When you declare that 
final liberation arises from the knowledge of the truth, do 
you mean that liberation ensues immediately upon this 
knowledge being attained ? " We reply, " No," for it is 
said in the second Nyaya aphorism, " Pain, birth, activity, 
faults, false notions, on the successive annihilation of 
these in turn, there is the annihilation of the one next 
before it," by means of this knowledge of the truth. Now 
false notions are the thinking the body, &c., which are 
not the soul, to be the soul ; " faults " are a desire for those 
things which seem agreeable to the soul, and a dislike to 
those things which seem disagreeable to it, 3 though in 
reality nothing is either agreeable or disagreeable to the 
soul. And through the mutual reaction of these different 
" faults " the stupid man desires and the desiring man is 
stupid ; the stupid man is angry, and the angry man is 
stupid. Moreover the man, impelled by these faults, does 
those things which are forbidden: thus by the body he does 
injury, theft, &c. ; by the voice, falsehood, &c. ; by the mind, 
malevolence, &c. ; and this same sinful " activity " pro 
duces demerit. Or, again, he may do laudable actions by 

1 Action (pravritti) follows after the 3 The printed text omits the third 
ascertainment of the truth by nydya. fault, "a stupid indifference, moha," 

2 Cp. Va"tsyayana s Comment., p. which is however referred to pre- 
6. The Calcutta edition reads pra- sently. 

kirtitd for parikshitd. 


his body, as alms, saving others, &c., truthful speaking, 
upright counsel, &c., by his voice, and guilelessness, &c., 
by his mind ; and this same right activity produces merit. 
But both are forms of activity, and each leads to a 
similar laudable or blamable birth or bodily manifesta 
tion ; and while this birth lasts there arises the impression 
of " pain," which we are conscious of as of something that 
jars against us. Now this series, beginning with "false 
notions" and ending with "pain," is continually going 
on, and is what we mean by the words " mundane exist 
ence," which rolls on ceaselessly, like a waterwheel. And 
whenever some pre-eminent man, by the force of his 
previous good deeds, obtains through the teaching of a 
great teacher the knowledge that all this present life is 
only a scene of pain and bound up with pain, he recognises 
that it is all to be avoided, and desires to abolish the 
ignorance, &c., which are the causes that produced it. 1 
Then he learns that the one means to abolish it is the 
knowledge of the truth; and as he meditates on the 
objects of right knowledge divided into the four sciences, 2 
there arises in his mind the knowledge of the truth, or, in 
other words, a right view of things as they are ; and from 
this knowledge of the truth false notions disappear. When 
false notions disappear, the " faults " pass away ; w r ith 
them ceases " activity ; " and with it ceases " birth ; " and 
with the cessation of " birth " comes the entire abolition 
of " pain," and this absolute abolition is final bliss. Its 
absoluteness consists in this, that nothing similar to that 
which is thus abolished can ever revive, as is expressly 
said in the second aphorism of the Nyaya Sutras : " Pain, 
birth, activity, faults, false notions, since, on the successive 
annihilation of these in turn, there is the annihilation of 

1 In p. 1 1 6, line 3, I would read the causes of the stability of the 

tannirvartakam for tannivartakam. world " (cf. Manu, vii. 43). It 

2 This refers to the couplet so occurs in Kamandaki s Nitisdra, ii. 

often quoted in Hindu authors, 2, and seems to be referred to in 

"Logic, the three Vedas, trade and Vdtsyayana s Com. p. 3, from which 

agriculture, and the eternal doctrine Mddhava is here borrowing, 
of polity, these four sciences are 


the one next before it, there is [on the annihilation of the 
last of them] final beatitude." 

"But is not your definition of the summum lonum, 
liberation, i.e., the absolute abolition of pain/ after all 
as much beyond our reach as treacle on the elbow is to 
the tongue ; 1 why then is this continually put forth as if 
it were established beyond all dispute ? " We reply that 
as all those who maintain liberation in any form do 
include therein the absolute abolition of pain, our defini 
tion, as being thus a tenet accepted in all the schools, 
may well be called the royal highway 2 of philosophy. 
No one, in fact, maintains that pain is possible without 
the individual s activity. Thus even the Madhyamika s 
opinion that "liberation consists in the abolition of soul," 
does not controvert our point, so far at any rate as that it 
is the abolition of pain. But if you proceed to argue that 
the soul, as being the cause of pain, is to be abolished just 
like the body, &c., we reply that this does not hold, since 
it fails under either alternative. For do you mean by 
" the soul," (a.) the continued succession of cognitions, or 
(&.) something different therefrom ? (a.) If the former, we 
make no objection, [since we Naiyayikas allow that cogni 
tion is evanescent, 3 and we do desire to abolish cognition 
as a cause of pravritti or action 4 ], for who would oppose 
a view which makes for his own side ? (&.) But if the 
latter, then, since it must be eternal, 5 its abolition is 
impossible ; and, again, a second objection would be that 
no one would try to gain your supposed summum lonum;" 
for surely no sensible person would strive to annihilate 
the soul, which is always the dearest of all, on the prin- 

1 Compare the English proverb, first moment, remains during the 
" As soon as the cat can lick her second, and ceases in the third. 
ear . * gee Nyciva Sut. i. 2. 

2 Literally the "bell- road," i.e., 5 As otherwise why should we 
" the chief road through a village, require liberation at all ? Or rather 
or that by which elephants, &c., the author probably assumes that 
decorated with tinkling ornaments, other Naiydyikas have sufficiently 
proceed." Wilson s Diet. established this point^ against its 

3 The cognition is produced in the opponents, cf. p. 167, line II. 


ciple that "everything else is dear for the soul s pleasure;" 
and, again, everybody uses such a phrase as " liberated/ 
[and this very term refutes the idea of annihilation or 

" But why not say with those Bauddhas who hold the 
doctrine of pure intelligence [i.e., the Yogacharas and the 
Sautrantikas 1 ], that the summum lonum is the rising of 
pure intelligence consequent on the cessation of the con 
scious subject ? " To this view we object that there is an 
absence of means ; and also it cannot be established that 
the locus [or subject] of the two states is the same. For 
the former, if it is replied that the well-known fourfold 
set of Bauddha contemplations 2 are to be accepted as the 
cause, we answer that, as [according to the Bauddha tenet 
of the momentary existence of all things] there cannot be 
one abiding subject of these contemplations, they will 
necessarily exercise a languid power like studies pursued 
at irregular intervals, and be thus ineffectual to produce 
any distinct recognition of the real nature of things. 

And for the latter, since the continued series of cogni 
tions when accompanied by the natural obstacles 3 is said 
to be " bound," and when freed from those obstacles is 
said to be " liberated," you cannot establish an identity 
of the subject in the two states so as to be able to say 
that the very same being which was bound is now 

Nor do we find the path of the Jainas, viz., that " Libera 
tion is the releasing from all obstructions, " a path en 
tirely free from bars to impede the wayfarer. Pray, will our 
Jaina friend kindly inform us what he means by " obstruc 
tion " ? 4 If he answers " merit, demerit, and error," we 
readily grant what he says. But if he maintains that 
" the body is the true obstruction, and hence Liberation is 
the continual upspringing of the soul consequent on the 

1 See supra, pp. 24-32. 3 In the form of the various Mesas 

2 All is momentary, all is pain, or "afflictions." 

all is sui generis, all is unreal. * Avarana, cf. pp. 55, 58. 


body s annihilation, as of a parrot released from its 
cage/ then we must inquire whether this said soul 
possesses form or not. If it possesses form, then has it 
parts or not? If it has no parts, then, since the well- 
known definition of an atom will apply here as "that 
which has form without parts," it will follow that the 
attributes of the soul are, like those of an atom, impercep 
tible to the senses. 1 If you say that it has parts, then 
the general maxim that " whatever has parts is non- 
eternal," would necessitate that the soul is non-eternal ; 
and if this were conceded, then two grand difficulties 
[against the Providential course of the world] would burst 
in unopposed, viz., that what the soul has done would, at 
its cessation, perish with it [and thus fail of producing 
the proper fruit], while it would have reaped during life 
the effects of what it had not done [as the good and evil 
which happened to it would not be the consequences of 
its actions in a former birth]. If, on the other hand, the 
Jaina maintains that the soul does not possess form at all, 
then how can he talk of the soul s " upspringing," since 
all such actions as motion necessarily involve an agent 
possessing form ? 2 

Again, if we take the Charvaka s view " that the only 
bondage is dependence on another, and therefore indepen 
dence is the true liberation," if by " independence" he 
means the cessation of pain, we have no need to controvert 
it. But if he means autocratic power, then no sensible 
man can concede it, as the very idea of earthly power 
involves the idea of a capability of being increased and of 
being equalled. 3 

Again, the Sankhya opinion, which first lays down that 
nature and soul are utterly distinct, and then holds that 

1 But the Nytiya holds that the is difficult, but I believe that prati- 
attributes of the soul, as happiness, landha means here vydpti, as it does 
desire, aversion, &c., are perceived in Sankhya Sutras, i. 100. 

by the internal sense, mind (Bhdsha 3 The true summum bonum must 
P- 83)- be niratisaya, incapable of being 

2 The reading murtapratibandMt added to. 


" liberation is the soul s remaining as it is in itself after 
nature [on being known] has withdrawn," even this 
opinion accepts our tenet of the abolition of pain; but 
there is left a difficulty as to whether this cognition of 
the distinction between nature and soul resides in the 
soul or in nature. It is not consistent to say that it 
resides in the soul, since the soul is held to be unchange 
able, and this would seem to involve that previously it 
had been hampered by ignorance ; nor can we say that it 
resides in nature, since nature is always held to be un 
intelligent. Moreover, is nature spontaneously active or 
inactive ? If the former, then it follows that there can be 
no liberation at all, since the spontaneous actions of things 
cannot be set aside ; and if the latter, the course of mun 
dane existence would at once cease to go on. 

Again, we have the same recognition of our " abolition 
of pain " in the doctrine of Bhatta Sarvajna and his 
followers, that " Liberation is the manifestation of an 
eternal happiness incapable of being increased ; " but here 
we have the difficulty that an eternal happiness does not 
come within the range of definite proof. If you allege 
Sruti as the proof, we reply that Sruti has no place when 
the thing itself is precluded by a valid non-perception ; l or 
if you allow its authority, then you will have to concede 
the existence of such things as floating stones. 2 

"But if you give up the view that liberation is the 
manifestation of happiness, and then accept such a view 
as that which holds it to be only the cessation of pain, 
does not your conduct resemble that of the dyspeptic 
patient who refused sweet milk and preferred sour rice- 
gruel?" Your satire, however, falls powerless, as fitter 
for some speech in a play [rather than for a grave philoso 
phical argument]. The truth is that all happiness must 

1 Yogydnupaldbdhi is when an "grdvdnah plavanti," see Uttara 
object is not seen, and yet all the Naishadha, xvii. 37. The phrase 
usual concurrent causes of vision are amdnah plavanti occurs in Shadv. 
present, as the eye, light, c. Br. 5, 12. 

2 Alluding to the Vedic phrase, 


be included under the category of pain, since, like honey 
mixed with poison, it is always accompanied by pain, 
either as admitting of increase, 1 or as being an object of 
perception, or as being exposed to many hostile influences, 
or as involving an irksome necessity of seeking all kinds 
of instruments for its production. Nor may you retort on 
us that we have fulfilled the proverb of "seeking one 
thing and dropping another in the search," since we have 
abolished happiness as being ever tainted by some inci 
dental pain, and, at the same time, our own favourite 
alternative is one which no one can consider desirable. 
For the truth is that any attempt to establish happiness 
as the summum lonum, since it is inevitably accompanied 
by various causes of pain, is only like the man who 
would try to grasp a red-hot ball of iron under the delusion 
that it was gold. In the case of objects of enjoyment got 
together by rightful means, we may find many firefly-like 
pleasures; but then how many are the rainy days to drown 
them ? And in the case of those got together by wrong 
means, the mind cannot even conceive the future issue 
which will be brought about. Let our intelligent readers 
consider all this, and not attempt to disguise their own 
conscious experience. Therefore it is that we hold it as 
indisputable that for him, pre-eminent among his fellows, 
who, through the favour of the Supreme Being, has, by 
the regular method of listening to the revealed Sruti, &c., 
attained unto the knowledge of the real nature of the soul, 
for him the absolute abolition of pain is the true Liberation. 
But it may be objected, " Is there any proof at all for 
the existence of a Supreme Being, i.e., perception, infer 
ence, or Sruti ? Certainly perception cannot apply here, 
since the Deity, as devoid of form, &c., must be beyond 
the senses. JSTor can inference hold, since there is no 
universal proposition or true middle term which can 
apply. 2 Nor can Sruti, since neither of the resulting 

1 Or perhaps capable of being surpassed." 

2 Since the Supreme Being is a single instance. 


alternatives can be sustained ; for is it supposed to reveal, 
as being itself eternal, or as non-eternal ? Under the former 
view an established tenet of our school would be con 
tradicted [viz., that the Veda is non-eternal] ; under the 
latter, we should be only arguing in a circle. 1 As for 
comparison and any other proof which might be adduced 
[as that sometimes called presumption, &c.], they need 
not be thought of for a moment, as their object matter 
is definitely limited, and cannot apply to the present case. 2 
Therefore the Supreme Being seems to be as unreal as a 
hare s horn." But all this elaborate disputation need excite 
no flurry in the breast of the intelligent, as it can be at 
once met by the old argument, "The mountain, seas, &c., 
must have had a maker from their possessing the nature 
of effects just like a jar." (a.) Nor can our middle term 
[possessing the nature of effects] be rejected as unproved 
(asiddha), since it can be established beyond a doubt by the 
fact of the subject s possessing parts. " But what are we to 
understand by this possessing parts ? Is it existing in 
contact with parts/ or in intimate relation with parts ? 
It cannot be the first, since this would equally apply to 
such eternal things as ether, 3 &c. ; nor can it be the 
second, since this would prove too much, as applying to 
such cases as the [eternal] species, thread, which abides 
in intimate relation with the individual threads. It there 
fore fails as a middle term for your argument." We reply, 
that it holds if we explain the "possessing parts" as 
" belonging to the class of those substances which exist in 
intimate relation." 4 Or we may adopt another view and 

1 Since the Veda, if non-eternal, tact with the parts of everything, as 
must [to be authoritative] have e.g., a jar. 

been created by God, and yet it 4 The whole (as the jar) resides 

is brought forward to reveal the by intimate relation in its parts (as 

existence of God. the jar s two halves). But the eter- 

2 The Nyaya holds presumption nal substances, ether, time, the soul, 
to be included under inference, and mind, and the atoms of earth, water, 
comparison is declared to be the fire, and air, do not thus reside in any- 
ascertaining the relation of a name thing, although, of course, the cate- 
to .the thing named. gory visesha does reside in them by 

3 Since ether is connected by con- intimate relation. The word " sub- 


maintain that it is easy to infer the " possessing the nature 
of effects " from the consideration of their possessing in 
termediate magnitude. 1 

(5.) Nor can our middle term be rejected as "con 
tradictory" (viruddhc?)* since there is no such acknow 
ledged universal proposition connected with it as would 
establish the opposite major term to that in our syllogism 
[i.e., that they must have had no maker], (c.) Nor is our 
middle term too general (anaikdnta), since it is never 
found in opposite instances [such as the lake, which is the 
mpaksha in the argument, " The mountain has fire because 
it has smoke "]. (d.} Nor again is it precluded (bddhita 
or kdldtyayopadislita), for there is no superior evidence to 
exercise such a precluding power. (e.~) Nor is it counter 
balanced (sat-pratipaJcshita), for there does not appear to 
be any such equally valid antagonist. 

If you bring forward as an antagonistic syllogism, 
" The mountains, &c., cannot have had a maker, from the 
fact that they were not produced by a body, just as is the 
case with the eternal ether," this pretended inference 
will no more stand examination than the young fawn can 
stand the attack of the full-grown lion ; for the additional 
words " by a body " are useless, since " from the fact that 
they were not produced" would be a sufficient middle 
term by itself [and the argument thus involves the fallacy 
called vydpyatvdsiddhi]? Nor can you retort, " Well, let 
this then be our middle term ; " for you cannot establish 
it as a real fact. Nor again is it possible to raise the 

stances" excludes tantutva, and "ex- older Naiydyikas maintained that 

isting in intimate relation" excludes the argument the mountain has fire 

ether, &c. because it has blue smoke, involved 

1 Intermediate between infinite the fallacy of vya"pyatva"siddhi, be- 
and infinitesimal, all eternal sub- cause the alleged middle term was 
stances being the one or the other. unnecessarily restricted (see Sid- 

2 The viruddha-hetu is that which dhantaMuktdv.p.77). The moderns, 
is never found where the major term however, more wisely consider it as 
1S< a harmless error, and they would 

* This and much more of the rather meet the objection by assert- 

whole discussion is taken from the ing that there is no proof to establish 

Kusumanjali, v. 2, and I extract my the validity of the assumed middle 

note on the passage there. "The term." 



smallest shadow of a fear lest our middle term should be 
liable to limitation by any suggested condition (upddhi), 1 
[such as "the being produced by a corporeal agent," to 
limit our old reason " from having the nature of effects "], 
because we have on our side a valid line of argument to 
establish our view, viz., " If the mountains, &c., had no 
maker, then they would not be effects " [but all do acknow 
ledge that they have the nature of effects], for in this world 
that is not an effect which can attain its proper nature in 
dependently of any series of concurrent causes. And this 
series inevitably involves the idea of some sort of maker ; 
and I mean by "being a maker" the being possessed of that 
combination of volition, desire to act, and knowledge of 
the proper means, which sets in motion all other causes, 
but is itself set in motion by none. And hence we hold 
that if the necessity of a maker were overthrown, the 
necessity of the action of all the other causes would be 
simultaneously overthrown, since these are dependent 
thereon ; and this would lead to the monstrous doctrine 
that effects could be produced without any cause at all. 
There is a rule laid down by Sankara-kinkara which 
applies directly to the present case 

" When a middle term is accompanied by a sound argu 
ment to establish its validity, 

" Then you cannot attempt to supply a limiting con 
dition on account of the [supposed] non-invariable 
concomitance of the major term." 

If you maintain that there are many sound counter 
arguments, such as " If the Supreme Being were a maker, 
He would be possessed of a body," &c., we reply, that all 
such reasoning is equally inconsistent, whether we allow 
that Supreme Being s existence to be established or not. 2 

1 For the upddhi cf. pp. 7, 8. 

2 As in the former case it would be 
clear that it is a subject for separate 
discussion ; and in the latter you 
would be liable to the fault of dsray- 

dsiddhi, a baseless inference," since neither can an unreal thin 
your subject (or minor term), being subject of a negation." 

itself non-existent., cannot be the 
locus or subject of a negation (cf. 
Kusumaiijali, iii. 2). "Just as that 
subject from which a given attribute 
is excluded cannot be unreal, so 
be the 


As has been said by Udayana Acharya [in the Kusuman- 
jali, iii % 5] 

" If Sruti, &c., have any authority, your negative argu 
ment fails from being precluded ; if they are falla 
cious, our old objection of a baseless inference 
returns stronger than ever." 

Nor need we fear the possibility of any other contra 
diction to our argument, since it would be overthrown by 
either alternative of God s being known or unknown. 1 

" Well, let all this be granted ; but the activity of God in 
creating the world, what end did it have in view ? His own 
advantage or some other being s ? If it was for the former 
end, was it in order to attain something desired, or to 
avoid something not desired ? It could not be the first, 
because this would be quite incongruous in a being who 
possesses every possible desire gratified ; and for the same 
reason too it could not be the second. If it was for the 
latter end [the advantage of another] it would be equally 
incongruous ; for who would call that being " wise " who 
busied himself in acting for another ? If you replied that 
His activity was justified by compassion, any one would at 
once retort that this feeling of compassion should have 
rather induced Him to create all living beings happy, and 
not checkered with misery, since this militates against 
His compassion ; for we define compassion as the disin 
terested wish to avoid causing another pain. Hence we 
conclude that it is not befitting for God to create the 
world. This has been said by Bhattacharya 

" Not even a fool acts without some object in view ; 

" Suppose that God did not create the world, what end 

would be left undone by Him ? " 
We reply, thou crest-jewel of the atheistic school, be 

1 If God is known, then His exis- pardkatatvdt, and then begin the 

tence must be granted ; if He is not next clause with sydd etat. The 

known, how can we argue about printed text, vikcdpapardhatah sydt 

Him? I read lines 15, 16, in p. tad etat, seems unintelligible. 
1 20 of the Calcutta edition, vikalpa- 


pleased for a moment to close thy envy-dimmed eyes, 
and to consider the following suggestions. His action in 
creation is indeed solely caused by compassion; but the 
idea of a creation which shall consist only of happiness is 
inconsistent with the nature of things, since there cannot 
but arise eventual differences from the different results 
which will ripen from the good or evil actions of the beings 
who are to be created. Nor need you object that this 
would interfere with God s own independence [as He 
would thus seem to depend on others actions], since there 
is the well-known saying, " One s own body does not 
hinder one ; " nay rather it helps to carry out one s aims ; l 
and for this there is authority in such passages of the 
Veda as that (in the Svetasvatara Upanishad, iii. 2), "There 
is one Eudra only; he admits 2 not of a second," &c. " But 
then how will you remedy your deadly sickness of reason 
ing in a circle ? [for you have to prove the Veda by the 
authority of God, and then again you have to prove God s 
existence by the Veda "]. We reply, that we defy you to 
point out any reasoning in a circle in our argument. Do 
you suspect this " reciprocal dependence of each," which 
you call " reasoning in a circle," in regard to their being 
produced or in regard to their being known ? 3 It cannot 
be the former, for though the production of the Veda is 
dependent on God, still as God Himself is eternal, there 
is no possibility of His being produced ; nor can it be in 
regard to their being known, for even if our knowledge 
of God were dependent on the Veda, the Veda might be 
learned from some other source ; nor, again, can it be in 
regard to the knowledge of the non-eternity of the Veda, 
for the non-eternity of the Veda is easily perceived by 

1 The aggregate of the various 2 The usual reading is tasthur for 

subtile bodies constitutes Hiranya- tasthe. 

garbha, or the supreme soul viewed 3 For these divisions of the anyon- 

in His relation to the world as creator, ydsraya fallacy, see Nydyasutra, vritti, 

while the aggregate of the gross i. 39 (p. 33). 
bodies similarly constitutes his gross 
body (viraj). 


any yogin endowed with the transcendent faculties (tivra, 1 

Therefore, when God has been rendered propitious by 
the performance of duties which produce His favour, the 
desired end, Liberation, is obtained ; thus everything is 
clear. E. B. C. 

NOTE ON PAGES 172, 173. 

We have here an exemplification of the five fallacies or hetvdbkdsas 
of the modern Hindu logic (cf. Siddhdntamuki., 71, Tarkasarngr., 
55~67), viz., anaifamta, viruddha, asiddha, kdldtyayopadishta or bd- 
dhita, and pratipaJcshita or sat-pratipaksha. The four first of these 
generally correspond to the savyabhichdra or " erratic," viruddha or 
"contradictory," sddhyasama or "unproved," and atitakdla or "mis 
timed," i.e., " precluded," as given in the list of fallacies of the older 
logic in p. 164 ; but pratipakshita corresponds imperfectly to praka- 
ranasama. The prakaranasama or " uncertain " reason is properly 
that reason which is equally available for both sides, as, e.g., the 
argument, " Sound is eternal because it is audible," which could be 
met by the equally plausible argument, "Sound is non-eternal be 
cause it is audible ; " or, according to other authorities, it is that 
reason which itself raises the same difficulties as the original ques 
tion, as, e.g., " sound is non-eternal because eternal qualities are not 
perceived in it ; " here this alleged reason is as much the subject of 
dispute as the old question, " Is sound eternal ? " But the pratipak 
shita reason is one which is counterbalanced by an equally valid 
reason, as " Sound is eternal because it is audible," and " Sound is 
non-eternal because it is a product." 

1 For tivra cf. Yoga sutras, i. 21, 22. 




AN objector may here ask, "Are you not continually 
repeating that merit (dharma) comes from the practice of 
duty (dharma), but how is duty to be denned or proved ? " 
Listen attentively to my answer. A reply to this ques 
tion has been given in the older 1 Mimamsa by the holy 
sa^-e Jaimini. Now the Mimamsa consists of twelve 

O * 

books. 2 In the first book is discussed the authoritativeness 
of those collections of words which are severally meant by 
the terms injunction (vidhi), "explanatory passage" (artha- 
vdda), hjimi (mantra), tradition (smritty, and "name." In 
the second, certain subsidiary discussions [as e.g., on ap&rva\ 
relating to the difference of various rites, refutation of 
(erroneously alleged) proofs, and difference of performance 
[as in "constant" and "voluntary" offerings]. In the third, 
Sruti, " sign " or " sense of the passage " (lingo), " con 
text" (y&kya), &c., and their respective weight when in 
apparent opposition to one another, the ceremonies called 
pratipattirkarmdni, things mentioned incidentally (andra- 
bkyddhitd), things accessory to several main objects, as 
praydjas, &c., and the duties of the sacrifices In the 
fourth, the influence on other rites of the principal and 
subordinate rites, the fruit caused by the juliu being 
made of the butea frondosa, &c., and the dice-play 
ing, &c., which form subordinate parts of the rdjasuya 
sacrifice. In the fifth, the relative order of different 

1 Mddhava here calls it the prdcki Mimamsa. 

2 Cf . /. Nydyamdldvist, pp. 5-9. 


passages of Sruti, &c., the order of different parts of a 
sacrifice [as the seventeen animals at the vdjapeya], the 
multiplication and non-multiplication of rites, and the 
respective force of the words of Sruti, order of mention, 
&c., in determining the order of performance. In the 
sixth, the persons qualified to offer sacrifices, their obliga 
tions, the substitutes for enjoined materials, supplies for 
lost or injured offerings, expiatory rites, the sattra offer 
ings, things proper to be given, and the different sacrificial 
fires. In the seventh, transference of the ceremonies of 
one sacrifice to another by direct command in the Yaidic 
text, and then as inferred by "name" or "sign." In the 
eighth, transference by virtue of the clearly expressed or 
obscurely expressed " sign," or by the predominant " sign," 
and cases where no transference takes place. In the 
ninth, the beginning of the discussion on the adaptation 
of hymns when quoted in a new connection (uha), the 
adaptation of sdmans and mantras, and collateral questions 
connected therewith. In the tenth, the discussion of 
occasions where the non-performance of the primary rite 
involves the " preclusion " and non-performance of the 
dependent rites, and of occasions where rites are precluded 
because other rites produce their special result, discussions 
connected with the graha offerings, certain sdmans, and 
various other things, and a discussion on the different 
kinds of negation. In the eleventh, the incidental mention 
and subsequently the fuller discussion of tantra I [where 
several acts are combined into one], and dvdpa [or the per 
forming an act more than once]. In the twelfth, a discus 
sion on prasanga [where the rite is performed for one chief 
purpose, but with an incidental further reference], tantra, 
cumulation of concurrent rites (samuchchaya) and option. 
Now the first topic which introduces the discussions of 

1 Thus it is said that he who de- tantra one offering to Agni would do 

sires to be a family priest should for both ; but as the offering to Soma 

offer a black-necked animal to Agni, comes between, they cannot be 

a parti-coloured one to Soma, and united, and thus it must be a case 

a black- necked one to Agni. Should of dvdpa, i.e., offering the two sepa- 

this be a case for tantra or not ? By rately (/. Nydyamdld, xi. i, 13). 


the Purva-Mimamsa arises from the aphorism, "Now there 
fore a desire to know duty [is to be entertained by thee"]. 
Now the learned describe a " topic " as consisting of five 
members, and these are (a.) the subject, (&.) the doubt, 
(c.) the primd facie argument, (d.) the demonstrated con 
clusion, and (e.) the connection (sangati). The topic is dis 
cussed according to the doctrines held by the great teachers 
of the system. Thus the " subject " to be discussed is the 
sentence, "The Veda is to be read." Now the "doubt" which 
arises is whether the study of Jaimini s sdstra concerning 
duty, beginning with the aphorism, " Duty is a thing which 
is to be recognised by an instigatory passage," and ending 
with " and from seeing it in the anvdhdrya" is to be com 
menced or not. The primd facie argument is that it is not 
to be commenced, whether the injunction to read the Yeda 
be held to have a visible and present or an invisible and 
future fruit, (a.) If you say that this injunction must have 
a visible fruit, and this can be no other l than the know 
ledge of the meaning of what is read, we must next ask 
you whether this said reading is enjoined as something 
which otherwise would not have been thought of, or 
whether as something which otherwise would have been 
optional, as we see in the rule for shelling rice. 2 It can 
not be the former, for the reading of the Veda is a means 
of knowing the sense thereof from its very nature as 
reading, just as in the parallel instance of reading the 
Mahabharata ; and we see by this argument that it would 
present itself as an obvious means quite independently 
of the injunction. Well, then, let it be the latter alterna 
tive ; just as the baked flour cake called puroddsa is made 
only of rice prepared by being unhusked in a mortar, 
when, but for the injunction, it might have been unhusked 
by the finger-nails. There, however, the new moon and full 
moon sacrifices only produce their unseen effect, which is 

1 In p. 123, line 4, I read vilak- the lines vidhir atyantam aprdpto 
sJmna-drislttaphala. niyamah pdkshike sati, tatra chdn- 

2 In the former case it would be a yatra cha prdptau parisainkliyd vidhi- 
vidki, in the latter a niyama, Cf. yate. 


the principal aptirva, by means of the various minor effects 
or subordinate aptirvas, produced by the various subordi 
nate parts of the whole ceremony ; and consequently the 
minor apurva of the unhusking is the reason there for the 
restricting injunction. But in the case which we are dis 
cussing, there is no such reason for any such restriction, 
as the rites can be equally well performed by gaining the 
knowledge of the Veda s meaning by reading a written 
book, or by studying under an authorised teacher. Hence 
we conclude that there is no injunction to study the Piirva 
Mimamsa as a means of knowing the sense of the Veda. 
(&.) "What, then, becomes of the Vedic injunction, The Veda 
is to be read ?" Well, you must be content with the fact 
that the injunction will have heaven as its [future] fruit, 
although it merely enjoins the making oneself master of the 
literal words of the Vedic text [without any care to under 
stand the meaning which they may convey], since heaven, 
though not expressly mentioned, is to be assumed as the 
fruit, according to the analogy of the Visvajit offering. Just 
as Jaimini, in his aphorism (iv. 3, 15), " Let that fruit be 
heaven, since it equally applies to all," establishes that 
those who are not expressly mentioned are still qualified 
to offer the Visvajit sacrifice, and infers by argument that 
its characteristic fruit is heaven, so let us assume it to be 
in the present case also. As it has been said 

" Since the visible fruit would be equally obtained with 
out the injunction, this cannot be its sole object ; we must 
rather suppose heaven to be the fruit from the injunction s 
significance, after the analogy of the Visvajit, &c." 

Thus, too, we shall keep the Smriti rule from being 
violated : " Having read the Veda, let him bathe." For this 
rule clearly implies that no long interval is to take place 
between reading the Veda and the student s return to his 
home ; while, according to your opinion, after he had read 
the Veda, he would still have to remain in his preceptor s 
house to read the Mimamsa discussions, and thus the idea 
of no interval between would be contradicted. Therefore 


for these three reasons, (a.) that the study of Mimamsa is 
not enjoined, (&.) that heaven can be obtained by the 
simple reading of the text, and (c.) that the rule for the 
student s return to his home is thus fulfilled, we maintain 
that the study of the Mimamsa discussions on duty is 
not to be commenced. 

The " authoritative conclusion " (siddhdnta), however, is 
as follows : 

We grant that it cannot be a case of vidhi, for it might 
have been adopted on other grounds ; but not even Indra 
with his thunderbolt could make us lose our hold of the 
other alternative that it is a case of niyama. In the sen 
tence, " The Veda is to be read," the affix tavya expresses 
an enforcing power in the word, 1 which is to be rendered 
visible by a corresponding action in man, bringing a certain 
effect into existence ; and this enforcing power seeks some 
corresponding end which is connected with the man s crea 
tive effort. Now it cannot be the act itself of reading, as 
suggested by the whole word adliyetavya, which it thus 
seeks as an end ; for this act of reading, thus expressed 
by the word, could never be regarded as an end, since it 
is a laborious operation of the voice and mind, consisting 
in the articulate utterance of the portion read. Nor could 
the portion read, as suggested by the whole sentence, be 
regarded as the end. For the mass of words called " Veda," 
which is what we really mean by the words " portion read," 
being eternal and omnipresent, could never fulfil the con 
ditions of the four " fruits of action," production, &c. 2 
Therefore the only true end which remains to us is the 

1 The Mimamsa holds that the make up a sacrifice possessing a cer- 
potential and similar affixes, which tain mystic influence ; " next it im- 
constitute a vidhi, have a twofold plies an enforcing power residing in 
power ; by the one they express an itself (as it is the word of the self- 
active volition of the agent, corre- existent Veda and not of God) which 
spending to the root-meaning (artha- sets the hearer upon this course of 
bhdvatld) ; by the other an enforcing action. 

power in the word (6abda-bhdvand). 2 These four "fruits of action" 
Thus in srargnkdmo yajeta, the eta are obscure, and I do not remember 
implies " let him produce heaven by to have seen them alluded to else- 
means of certain acts which together where. I was told in India that 


knowledge of the meaning, as obtained by carrying out the 
sense of the words of the injunction. According to the old 
rule, " He has the right who has the want, the power, and 
the wit," those who are aiming to understand certain things, 
as the new and full moon sacrifices, use their daily reading 
to learn the truth about them. And the injunction for read 
ing, since it virtually excludes the reading of written books, 
&c. [from the well-known technical sense of the word 
" read " when used in this connection], conveys the idea 
that the reading the Veda enjoined has a consecrated 
character [as taught by a duly authorised teacher]. There 
fore, as the principal apurva, produced by the great new 
and full moon sacrifices, necessitates and establishes the 
subordinate apurvas produced by the inferior sacrificial 
acts, as unhusking the rice, &c., so the mass of apurva 
produced by all the sacrifices necessitates and establishes 
a previous apurva produced by the restricting injunction 
(niyama), which prescribes reading the Veda as the means 
to know how to perform these sacrifices. If you hesitate 
to concede that a niyama could have this future influence 
called apurva, the same doubt might equally invalidate 
the efficacy of a vidhi [as the two stand on the same level 
as to their enjoining power]. Nor is the supposition a 
valid one that heaven is the fruit, according to the analogy 
of the Visvajit offering, since, if there is a present and 
visible fruit in the form of a knowledge of the meaning of 
the sacred text, it is improper to suppose any other future 
and unseen fruit. Thus it has been said 

" Where a seen fruit is obtained, you must not suppose 
an unseen one ; but if a vidhi has the restricting 
meaning of a niyama, it does not thereby become 

they were a thing s coining into ate, viparinamate, apalcsliiyate, nas- 

being, growing, declining, and per- yati. I do not see how there could 

ishing. If so, they are the second, be any reference to the four kinds 

third, fifth, and sixth of, the six of apurva, sc. pliala, samuddya, ut- 

viMras mentioned in Saiikara s patti, and ariga, described in Nyaya 

Vajrasiichi, 2, i.e., asti, jdyate, vardh- M. V. ii. I, 2. 


But an objector ma} 7 say, " Although a man who reads 
the simple text of the Yeda may not attain to a know 
ledge of its meaning, still, as he who reads the Yeda with 
its angaSj grammar, &c., may attain to this knowledge, the 
study of Mimamsa will be useless." But this is not true : 
for even though he may attain to a simple knowledge of 
the literal meaning, all deeper investigation must depend 
on this kind of discussion. For instance, when it is said, 
" He offers anointed gravel," neither grammar nor nigama 1 
nor ninikta will determine the true meaning that it is to 
be anointed with ghee and not with oil, &c. ; it is only by 
a Mimamsa discussion that the true meanino- is unravelled 


from the rest of the passage, " Yerily, ghee is brightness." 2 
It is therefore established that the study of Mimamsa is 
enjoined. Nor need it be supposed that this contradicts 
the passage of Smriti, "Having read the Yeda, let him 
bathe," which implies that he should now leave his teacher s 
house, and prohibits any further delay ; as the words do 
not necessarily imply that the return to the paternal roof 
is to follow immediately on his having read the Yeda, but 
only that it is to follow it at some time, and that both 
actions are to be done by the same person, just as we see 
in the common phrase, " Having bathed, he eats." There 
fore from the purport of the injunction we conclude that 
the study of the Piirva Mimamsa Sastra, consisting of a 
thousand "topics," 3 is to be commenced. This topic is 
connected with the main subject of the Sastra as being a 
subsidiary digression, as it is said, " They call that a subsi 
diary digression which helps to establish the main subject." 4 

I now proceed to give a sketch of the discussion of the 
same " topic " in accordance with the teaching of the Guru 

In the Smriti rule, 5 " Let him admit as a pupil the 
Brahman lad when eight years old (by investing him with 

1 The niyamas are the Vedic 4 This is to explain the last of the 
quotations in Ya"ska 3 nirukta. five members, the samgati. 

2 See Nyaya-mala-vistara, i. 4, 19. 5 Cf. AS valay ana s Grihya Sutras, 

3 The exact number is 915. i. 19, i. 


the sacred cord), let him instruct him," the object of the 
direction appears to be the pupil s instruction. Now a direc 
tion must have reference to somebody to be directed; and if 
you ask who is here to be directed, I reply, "He who desires 
to be a teacher," since, by Panini s rule (i. 3, 36), the root ni 
is used in the dtmancpada when honour, &c., are implied, i.e., 
here the duty which a teacher performs to his pupils. He 
who is to be directed as to admitting a pupil is the same 
person who is to be directed as to teaching him, since both 
are the object of one and the same command. Hence the 
inspired sage Mann has said (ii. 140), "The Brahman who 
girds his pupil with the sacrificial cord and then instructs 
him in the Veda, with its subsidiary aiigas and mystic 
doctrines, they call a spiritual teacher (dclidrya)" Now 
the teaching which is the function of the teacher cannot 
be fulfilled without the learning which is the function of 
the pupil, and therefore the very injunction to teach im 
plies and establishes a corresponding obligation to learn, 
since the influencer s efforts fail without those of one to be 
influenced. If you object that this view does not make 
reading the Veda the object of definite injunction, I reply, 
What matters it to us if it is not ? For even if there is 
no reason for us to admit a separate injunction for reading 
the Veda, it will still remain perpetually enjoined as a 
duty, because the passage which mentions it is a perpetual 
anuvdda or " supplementary repetition." x Therefore the 
former primd facie argument and its answer, which were 
given before under the idea that there was a definite 
injunction to read the Veda, must now be discussed in 
another way to suit this new view. 

Now the primd facie argument was that the study of 
Mimamsa, not being authoritatively enjoined, is not to be 
commenced ; the " conclusion " was that it is to be com 
menced as being thus authoritatively enjoined. 

1 The anuvdda, of course, implies anuvdda in the present case is the 

a previous vidhi, which it thus re- passage which mentions that the 

peats and supplements, and so carries Veda is to be read, as it enforces 

with it an equal authority. The the previous vidhi as to teaching. 


Now the upholders of the former or primd, facie view 
argue as follows : " We put to the advocates of the con 
clusion the following dilemma: Does the injunction to 
teach imply that the pupil is to understand the meaning 
of what is read, or does it only refer to the bare reading ? 
It cannot be the former, for obviously the act of teaching 
cannot depend for its fulfilment on the pupil s understand 
ing what is taught [as this will depend on his ability as a 
recipient]; and the latter will not help you, as, if the bare 
reading is sufficient, the Mimamsa discussions in question 
will have no subject or use. For their proper subject is a 
point in the Yeda, which is doubted about from having 
been only looked at in a rough and impromptu way ; now 
if there is no need of understanding the meaning at all, 
why should we talk of doubts and still more of any hope 
of ascertaining the true meaning by means of laborious 
discussion ? And therefore in accordance with the well- 
known principle, That which is a thing of use and not a 
matter of doubt is an object of attainment to an intelligent 
man, as, for instance, a jar which is in broad light and in 
contact with the external and internal senses, as there is 
in the present case no such thing as a subject to exercise 
it upon, or a useful end to be attained by it, we maintain 
that the study of Mimamsa is not to be commenced." 

We grant, in reply, that the injunction to teach does 
not imply a corresponding necessity that the student must 
understand the meaning ; still when a man has read the 
Veda with its subsidiary aiigas, and has comprehended 
the general connection of the words with their respective 
meanings, this will imply an understanding of the mean 
ing of the Yeda, just as it would in any ordinary human 
compositions. "But may we not say that, just as in 
the case of the mother who said to her son, Eat poison, 
the meaning literally expressed by the words was not 
what she wished to convey, since she really intended to 
forbid his eating anything at all in such and such a house ; 
so if the literal meaning of the Yeda does not express its 


real purport, the old objection will recur with full force 
that the study of Mimamsa will have neither subject nor 
end [as there will be no use in understanding the literal 
meaning, since, as in the mother s case, it may only lead 
astray, and so common sense must be the ultimate judge "]. 
We reply, that your supposed illustration and the case 
in question are not really parallel. In the supposed 
illustration the primary meaning of the words would 
be obviously precluded, because a direction to eat poison 
would be inconceivable in the mouth of an authoritative 
and trustworthy speaker like a mother, and you would 
know at once that this could not be what she wished to 
say ; but in the case of the Veda, which is underived from 
any personal author, why should not the literal meaning 
be the one actually intended ? And it is just the doubts 
that arise, as they occasionally will do, in reference to this 
intended meaning, which will be the proper " subject " of 
Mimamsa discussion ; and the settlement of these doubts 
will be its proper " end." Therefore, whenever the true 
meaning of the Veda is not obtained l by that reading 
which is virtually prescribed by the authoritative injunc 
tion to a Brahman to teach, it will be a proper subject for 
systematic discussion ; and hence we hold that the study 
of Mimamsa is enjoined, and should be commenced. 

" Well, 2 be it so " [say the followers of the Nyaya], " but 
how can the Yedas be said to be underived from any personal 
author, when there is no evidence to establish this ? 
Would you maintain that they have no personal author be 
cause, although there is an unbroken line of tradition, there 
is no remembrance of any author, just as is the case with 
the soul " ? 3 This argument is weak, because the alleged 
characteristics [unbroken tradition, &c.] are not proved; 
for those who hold the human origin of the Vedas main- 

1 I read in p. 127, line 12, anava- Dr. Muir s translation in his Sanskrit 
gamyamdnasya, and so the recension Texts, vol. iii. p. 88. 

given in the Nyaya M. V. p. 14, 3 The soul may be traced back 

na budhyamdnasya. through successive transmigrations, 

2 In the next two or three pages but you never get back to its begin- 
I have frequently borrowed from ning. 



tain that the line of tradition was interrupted at the time 
of the dissolution of the universe. And, again, what is 
meant by this assertion that the author is not remembered? 
Is it (i.) that no author is believed, or (2.) that no author 
is remembered ? The first alternative cannot be accepted, 
since we hold that God is proved to have been the author. 
Nor can the second, because it cannot stand the test of the 
following dilemma, viz., is it meant (a.) that no author of 
the Veda is remembered by some one person, or (6.) by any 
person whatever ? The former supposition breaks down, 
as it would prove too much, since it would apply to such 
an isolated stanza as " He who is religious and has over 
come pride and anger," &C. 1 And the latter supposition is 
inadmissible, since it would be impossible for any person 
who was not omniscient to know that no author of the 
Veda was recollected by any person whatever. Moreover, 
there is actual proof that the Veda had a personal author, 
for w r e argue as follows : The sentences of the Veda must 
have originated from a personal author, since they have 
the character of sentences like those of Kalidasa and other 
writers. And, again, the sentences of the Veda have been 
composed by a competent person, since, while they possess 
authority, they have, at the same time, the character of 
sentences, like those of Manu and other sages. 

But [ask the Mimamsakas] may it not be assumed that 
" all study of the Veda was preceded by an earlier study 
of it by the pupil s preceptor, since the study of the Veda 
must always have had one common character which was 
the same in former times as now ; " and therefore this un 
interrupted succession has force to prove the eternity of 
the Veda ? This reasoning, however [the Naiyayikas 

1 Madhava means that the author 
of this stanza, though unknown to 
many people, was not necessarily 
unknown to all, as his contempo 
raries, no doubt, knew who wrote it, 
and his descendants might perhaps 
still be aware of the fact. In this 
case, therefore, we have an instance 
of a composition of which some per- 

sons did not know the origin, but 
which, nevertheless, had a human 
author. The stanza in question is 
quoted in full in Bohtlingk s In- 
dische Spriiche, No. 5598, from the 
MS. anthology called the Sulhdshi- 
tdrnava. For muktaka, see Sdh. 


answer], cannot rise to the height of proof, for it has no 
more validity than such obviously illusory reasoning, as 
" All study of the Mahabharata was preceded by an earlier 
study of it by the pupil s preceptor, since it is the study 
of the Mahabharata, which must have been the same in 
former times as now." But [the Mimamsakas will ask 
whether there is not a difference beween these two cases, 
since] the Smriti declares that [Vishnu incarnate as] Vyasa 
was the author of the Mahabharata, in accordance with 
the line, " Who else than the lotus-eyed Yishnu could be 
the maker of the Mahabharata ? " [while nothing of this 
sort is recorded in any Smriti in regard to the Veda]. This 
argument, however, is pithless, since those words of the 
Purushasukta (Rig V., x. 90), " From him sprang the Rich 
and Saman verses ; from him sprang the Metres ; from him 
the Yajus arose; " prove that the Yeda had a maker. 

Further [proceed the ISTaiyayikas] we hold that sound 
is non-eternal l because it has genus, and is also percep 
tible to the external organs of beings such as ourselves, 
just as a jar is. 2 "But," you may object, "is not this 
argument refuted by the proof arising from the fact that 
we recognise the letter g (for example) as the same we 
have heard before?" This objection, however, is extremely 
weak, for the recognition in question is powerless to refute 
our argument, since it has reference only to identity of 
species, as in the case of a man whose hair has been cut 
and has grown again, or of a jasmine which has blossomed 
afresh. " But [asks the Mimamsaka] how can the Yeda 
have been uttered by the incorporeal Paramesvara, who 
has no palate or other organs of speech, and therefore 
cannot have pronounced the letters ? " " This objection 

1 The eternity of the Veda de- senses. Genera are themselves eter- 
pends on this tenet of the Mimdmsd nal (though the individuals in which 
that sound is eternal. they reside are not), but they have 

2 Eternal things (as the atoms of not themselves genus. Both these 
earth, fire, water, and air, minds, arguments belong rather to the 
time, space, ether, and soul) have Nydya-vaiseshika school than to the 
visesha, not sdmdnya or genus, and Nyaya. 

they are all imperceptible to the 

1 9 o 


[answers the Naiyayika] is not happy, because, though 
Paramesvara is by nature incorporeal, he can yet assume 
a body in sport, in order to show kindness to his wor 
shippers. Consequently the arguments in favour of the 
doctrine that the Yeda had no personal author are in 

I shall now [says the Mfmamsaka] clear up the whole 
question. What is meant by this paurusheyatva [" deri 
vation from a personal author "] which it is sought to 
prove? Is it (i.) mere procession (utpannatva) from a 
person, like the procession of the Veda from persons such 
as ourselves, when we daily utter it ? or (2.) is it the 
arrangement with a view to its manifestation of know 
ledge acquired by other modes of proof, as in the case of 
treatises composed by persons like ourselves ? If the first 
meaning be intended, there will be no dispute between 
us. 1 If the second sense be meant, I ask whether it is 
established (a.) by inference, 2 or (5.) by supernatural testi 
mony ? (a.) The former alternative cannot be correct, be 
cause your argument would equally apply to the sentences 
in dramas such as the Malatimadhava [which, of course, 
being a work of fiction, has no authoritative character]. 
If you qualify your argument by inserting the saving 
clause, "while they possess authority," 3 [as supra, p. iSS, 
line 21], even this explanation will fail to satisfy a philo 
sopher. For the sentences of the Veda are universally 
defined to be sentences which prove things that are not 
provable by other evidence. But if you could establish 
that these Vedic sentences only prove what is provable 
by other evidence, this definition would be at once con- 

1 The Mimdmsaka allows that the 
uchchdrana or utterance is non- 

2 The inference will be as follows : 
" The Vedas were arranged after 
being acquired by other modes of 
proof, with a view to their manifes 
tation, from the very fact of their 
having the nature of sentences, just 

like the compositions of Manu, 

3 The argument will now run, 
" The Vedas were arranged after 
being acquired by other modes of 
proof, because, while they possess 
authority, they still have the nature 
of sentences, like the composition of 
Manu, &c." 


tradicted, just as if a man were to say that his mother 
was a barren woman. And even if we granted that Para- 
mesvara might assume a body in sport, in order to show 
kindness to his worshippers, it would not at all follow 
that he would perceive things beyond the reach of the 
senses, from the want of any means of apprehending 
objects removed from him in place, in time, and in nature. 1 
Nor is it to be assumed that his eyes and other senses 
alone would have the power of producing such knowledge, 
for we can only draw upon our imagination in accordance 
with our past experience. This has been declared by the 
Guru [Prabhakara] when he refutes the supposition of an 
omniscient author 

" Wherever we do find the power of an organ intensified, 2 
it is done without its going beyond its own proper 
objects ; thus it may appear in the power of seeing 
the very distant or the very minute, but not in the 
ear s becoming cognisant of form." 

Hence (&.) we also maintain that your position cannot 
be established by any supposed supernatural testimony 
[as that quoted above from the Big- Veda, " from him 
sprang the Rich and Saman verses"]. For the rule of 
Panini (iv. 3, 101) will still remain inviolate, that the 
grammatical affixes with which such names as Kathaka, 
Kalapa, and Taittiriya are formed , impart to those deri 
vatives the sense of "uttered by" Katha, Kalapin, &c., 
though we maintain that these names have reference [not 
to those parts of the Veda as first composed by these 
sages, but] to the fact that these sages instituted certain 
schools of traditional study. And in the same way we 
hold [in reference to t)iis verse from the Big- Veda] that 
it only refers to the institution of certain schools of tra 
ditional study of these Vedas. 

Nor will any supposed inference establish the non- 

1 In assuming a material body, he 2 The Jainas allow thirty- four 
would be subject to material liniita- such superhuman developments (ati- 
tions. say ah] in their saints. 


eternity of sound, because [as we said before] it is opposed 
to the evidence of our consciousness, [since we certainly 
recognise the letter now heard as the one heard before]. 
Nor is it reasonable to reply that, although the letters are 
not the same, they seem to be so on account of their 
identity of species. For here we ask our opponents a 
question Is this idea that " the apparent sameness arises 
from identity of species 5 put forward from a wish to 
preclude entirely any idea of the letters being the same, 
or only [from an imagined fear of error] because experi 
ence shows that the recognition will sometimes be erroneous 
[as in the cases of the hair and jasmine mentioned above] ? 
(a.) If it arises from the latter reason, we Mimamsakas, 
who hold that the Veda is its own evidence, have said in 
reference to this timid imagination 

"He who foolishly imagines that something as yet 
unknown to him will come hereafter to stop his 
present conclusion, will go to utter ruin in every 
transaction of life, his mind a mass of doubts." 
(&.) "But [the Naiyayikas will ask] does not this recog 
nition of g and other letters [as the same which we heard 
before] refer to the species which exists the same in each, 
and not to the several individual letters, since, in fact, we 
perceive that they are different as uttered by different 
persons, otherwise we could not make such distinctions 
as we do when we say Somasarman is reading ? " This 
objection, however, has as little brilliancy as its prede 
cessors, for as there is no proof of any distinction between 
the individual g s, there is no proof that we ought to 
assume any such thing as a species g; and we maintain 
that, just as to the man who does not understand [the 
JSTaiyayika doctrine of] the species g, the one species [in 
the Naiyayika view] will by the influence of distinction of 
place, magnitude, form, and individual sounds, appear as 
if it were variously modified as itself distinct in place, as 
small, as great, as long, as short ; so to the man who does 
not understand our [Mimamsaka doctrine of] one individual 


g, the one g (in our view) will by the diversity of " mani- 
f esters," l appear to him associated with their respective 
peculiarities ; and as contrary characters are in this way 
ascribed [to the letter g], there is a fallacious appearance 
of distinction [between different g s]. But does this ascrip 
tion of contrary characters, which is thus regarded as 
creating a difference [between the g$], result (i.) from the 
nature of the thing, or (2.) from our imagination ? There 
is no proof of the former alternative ; for, if it were true, 
as an inherent difference would have to be admitted be 
tween different gs, we should have to say, " Chaitra has 
uttered ten g s" and not " Chaitra has uttered the same 
g ten times." On the latter supposition, there is no proof 
of any inherent distinction between gs, for inherent one 
ness is not destroyed by a difference of external disguises. 
Tims we must not conceive, from the apparent distinction 
caused by such external disguises as jars, &c., that there 
is any inherent distinction, as of parts, in the one indivi 
sible ether. The current use of the rejected phrase [i.e., 
" different " as applied to the g s] is really caused by the 
noise, which in each case is different. This has been said 
by the great teacher 

" The object which the Naiyayikas seek by supposing a 
species is, in fact, gained from the letter itself; 
and the object which they aim at by supposing an 
individuality in letters, is attained from audible 
noises; 2 so that the assumption of species is 
And again 

" Since in regard to sounds such an irresistible instinct 
of recognition is always awake within us, it pre 
cludes by its superior evidence all the inferences to 
prove sound s non-eternity." 
This at once refutes the argument given in the [Naiya- 

1 Jaimini maintains that the vibra- is these conjunctions and disjunc 
tions of the air "manifest" the al- tions, occasioned by the vibrations 
ways existing sound. of the air." Ballantyne, Mimdmsd 

" "What is meant by noise (ndda) Aphorisms, i. 17. 




yika] treatise by Vagiswara, entitled Mdna-manohara, 
" sound is non-eternal from the fact of its being a special 
quality belonging to an organ of sense 1 (sc. the ear), just 
as colour is to the eye." 

We can also refute it in the following ways : (a.) If we 
follow the [Sankhya and Vedanta] view that sound is a 
substance, it is evidently overthrown 2 [as in that case 
sound cannot be a quality] ; (&.) if we take it as referring 
to the noise, not the sound, we have no dispute, as it only 
establishes what we ourselves allow; and (c.) the infer 
ence is overthrown by the " limiting condition " [upddhi] 
of asrdvanatva, or " the not causing audition." 3 So Uda- 
yana tries at great length to establish that, although ether, 
the site of sound, is imperceptible, the non-existence of 
that which abides in this site is perceptible ; and he then 
brings forward as an evidence for the non- eternity of 
sound, that sense perception which causes the use of such 
common expressions as "The tumult is stopped," "The 
sound has arisen." 4 But he is sufficiently answered 5 by 
our old reply [in p. 193], that the fallacious appearance of 

1 The Nyaya holds that colour and 
sound are respectively special quali 
ties of the elements light and ether ; 
and as the organs of seeing and 
hearing are composed of light and 
ether, each will, of course, have its 
corresponding special quality. 

2 In p. 131, line 7, I read pro- 
tyalcslid siddheh. 

3 Cf . my note pp. 7, 8, (on the Char- 
vdka-dar.sana) for the upddhi. The 
upadlii or " condition " limits a too 
general middle term ; it is defined 
as " that which always accompanies 
the major term, but does not always 
accompany the middle." Thus if 
the condition "produced from wet 
fuel" is added to "fire," the argu 
ment " the mountain has smoke be 
cause it has fire " is no longer a false 
one. Here, in answer to the Nyaya 
argument in the text, our author 
objects that its middle term (" from 
the fact of its being a special quality 
belonging to an organ of sense ") 

is too wide, i.e., it is sometimes found 
where the major term " non-eternal 
is not found, as, e.g., in sound itself, 
according to the Mima msa doctrine. 
To obviate this he proposes to add the 
" condition," " not causing audition," 
as he will readily concede that all 
those things are non-eternal which, 
while not causing audition, are special 
qualities belonging to an organ of 
sense, as, e.g., colour. But I need 
scarcely add that this addition would 
make the whole argument nugatory. 
In fact, the Piirva Mimamsd, and the 
Nyaya can never argue together on 
this question of the eternity of sound, 
as their points of view are so totally 

4 In the former case we have the 
dliu-amsa of sound, in the latter its 

6 In p. 131, line 12, I read sama- 
pauhi for samdpohi, i.e., the passive 
aorist of sam 4- apa + ith. 


distinction arises from contrary characters being errone 
ously ascribed, just as, in the story, the demon Tala went 
away [as well as Betala] when the offering of blood was 
given to the latter. 1 And as for the objection raised by the 
author of the Nydyabliushana? that, if sound were eternal, 
the conclusion must follow that it would be either always 
perceptible or always imperceptible, this also is obviated 
by our allowing that we only perceive that sound which 
is manifested by our articulate noise. 3 And as for the 
(Naiyayika) argument against the existence 4 of such a 
constant relation as this which is supposed between the 
manifested " sound " and the manifesting " noise," since 
they both come simultaneously in contact with the sense 
of hearing, this is invalid, as it will indisputably apply 
with equal force in the case of the soul. 5 

Therefore as the Veda is thus proved to have not 
originated from any personal author, and as the minutest 
germ of suspicion against it is thus absolutely destroyed, 
we hold it as satisfactorily demonstrated that it has a 
self - established authority in all matters relating to 

" Well " 6 [say our opponents], " let this question rest ; 

1 I do not know this legend. Tala The Naiyayika argument would 

and Betala are the two demons who seem to be something as follows : 

carry Vikramaditya on their shoulders Sound is not thus manifested by 

in the Simhasan-battisi. It appears noise, since both are simultaneously 

to be referred to here as illustrating perceived by the senses, just as we see 

how one answer can suffice for two in the parallel case of the individual 

opponents. and its species ; these are both per- 

a This is probably a work by BM- ceived together, but the individual is 

sarvajna (see Dr. Hall s Bibl. Index, not manifested by the species. But 

p. 26). the Mimamsa rejoins that this would 

3 Dhvani, or our " articulate equally apply to the soul and know- 
noise," produces the vibrations of ledge ; as the internal sense perceives 
air which render manifest the ever- both simultaneously, and therefore 
existing sound. There is always an knowledge ought not to be mani- 
eternal but inaudible hum going on, fested by the soul, which is contrary 
which we modify into a definite to experience. But I am not sure 
speech by our various articulations, that I rightly understand the argu- 
I take samskrita here as equivalent ment. 

to abkivyakta. 6 Here begins a long piirvapcikslia, 

4 I read in p. 131, line i$,sarnskd- from p. 131, line 18, down to p. 133, 
rakasamskdryabhdvdbhdvdnumdnam. line 9 ; see p. 198 infra. 

5 It would be a case of vyabliichdra. 


but how about another well-known controversy ? It is 

" The Sankhyas hold that both authoritativeness and 
non-authoritativeness are self-proved; the followers of 
the Nyaya hold that both are proved by something else 
[as inference, &c.] ; the Buddhists hold that the latter is 
self-proved and the former proved by something else ; the 
teachers of the Yeda maintain that authoritativeness is 
self-proved and non-authoritativeness proved by some 
thing else. Now we ask, amidst all this discussion, how 
do the Mimamsakas accept as established their tenet that 
the authoritativeness of duty is self-proved ? And what 
is the meaning of this so-called self-proved authoritative- 
ness ? Is it (a.) that authoritativeness springs from itself ? 
or (&.) that it springs from the right knowledge in which 
it resides ? or (c.) that it springs from the instrumental 
causes [as the eye, &c.] which produced the right know 
ledge in which it resides ? or (rf.) that it resides in a par 
ticular knowledge produced by the instrumental causes 
which produced the right knowledge ? l or (e.) that it 
resides in a particular knowledge produced by the instru 
mental causes only which produced the right knowledge ? 

" (a.) It cannot be the first, because wherever the relation 
of cause and effect is found there must be a difference, 
and therefore these two cannot reside in the same subject 
[i.e., authoritativeness cannot cause itself]. (&.) It cannot 
be the second, because if knowledge, which is a quality, 
were the cause of authoritativeness, it would have to be a 
substance, as being an intimate cause. 2 (c.) It cannot be 
the third, because authoritativeness cannot properly be 

1 This is PrabMkara s view (see ~ Substances are "intimate causes" 
Siddh. Muktav., p. 118). The first to their qualities, and only substances 
knowledge is in the! form "This is a have qualities ; now if authoritative- 
jar ;" the second knowledge is the ness, which is a characteristic of right 
cognition of this perception in the knowledge, were caused by it, it 
form "I perceive the jar;" and this would be a quality of it, that is, 
latter produces authoritativeness right knowledge would be its inti- 
(prdmdnya), which resides in it as mate cause and therefore a sub- 
its characteristic. stance. 


produced at all, 1 whether we call it a general character 
istic (upddhi) or a species (jdti) ; 2 for if we call it an 
upddhi, it is defined as the absolute non-existence of any 
contradiction to a certain kind of knowledge which does 
not possess the nature of recollection ; 3 and this cannot be 
produced, for we all allow that absolute non-existence is 
eternal ; and still less can we speak of its being produced, 
if we regard it as a species, (d.) Nor can it be the fourth, 
for wrong knowledge [as well as right knowledge] is a par 
ticular kind of knowledge, and the instrumental causes 
which produce the general are included in those which pro 
duce the particular, 4 just as the general idea seed, as applied 
to tree, is included in the particular seed of any special 
tree, as, e.g., the Dalbergia Sisu ; otherwise we might sup 
pose that the particular had no instrumental cause at all. 
Your definition would therefore extend too far [and include 
erroneous as well as true knowledge] ; for non-authoritative- 
ness, which Yedantists and most Mimamsakas allow to be 
produced by something external, must also be considered 
as residing in a particular knowledge [i.e., a wrong know 
ledge] produced [in part] by the instrumental causes which 
produced the right knowledge. (e.) As for your fifth 
view, we ask whether by being produced by the instru 
mental causes only which produced right knowledge, you 
mean to include or exclude the absence of a defect ? It 
cannot be the former alternative ; because the followers of 
the Nyaya who hold that authoritativeness is proved by 
something external [as inference, &c.], would at once grant 
that authoritativeness is produced by the instrumental 
causes of knowledge combined with the absence of a defect. 

1 The eye, &c., would be its in- 3 The Purva Mima*msd, denies that 
strumental causes. recollection is right knowledge. 

2 The first three categories " sub- 4 Wrong knowledge is produced 
stance," " quality," and " action," by the same instrumental causes (as 
are called jdtis or species ; the last the eye, &c.) which produced right 
four, "genus," "visesha," "intimate knowledge, but by these together with 
relation," and "non-existence," are a "defect," as biliousness, distance 
called upddhis or "general charac- &c. 



Neither can it be the latter alternative ; for, inasmuch as 
it is certain that the absence of a defect is found com 
bined with the various instrumental causes, this absence of 
a defect is fixed as by adamantine glue to be a cause of 
right knowledge, since right knowledge will always ac 
company its presence, and be absent if it is absent, 1 and 
it will at the same time be not an unimportant condition. 2 
If you object that non-existence (or absence) cannot be a 
cause, we reply by asking you whether non-existence can 
be an effect or not ? If it cannot, then we should have to 
allow that cloth is eternal, as its " emergent non-existence" 
or destruction would be impossible. If it can be an effect, 
then why should it not be a cause also? So this rope 
binds you at both ends. This has also been said by Uda- 
yana [in his Kusumafijali, i. 10] 

" As existence, so too non-existence is held to be a cause 
as well as an effect. 

" The argument, in my opinion, runs as follows : Eight 
knowledge depends on some cause 3 other than the common 
causes of knowledge, from the very fact that, while it is an 
effect, it is also knowledge, just as wrong knowledge does. 4 
Authoritativeness is known through something external to 
itself [e.g., inference], because doubt arises in regard to it in 
an unfamiliar case, as we also see in non-authoritativeness. 

"Therefore, as we can prove that authoritativeness is 
both produced and recognised by means of something 
external, the Mimamsa tenet that authoritativeness is 
self-proved is like a gourd overripe and rotten." 

This long harangue of our opponent, however, is but a 
vain attempt to strike the sky with his fist ; for (a.) we 
mean by our phrase " self-proved " that while right know 
ledge is produced by the instrumental causes of know- 

1 Scil. if there be doshdbkdva there 3 Scil. or the absence of " defect," 
hpramd; if not, not. In p. 132, line doshdbhdva. 

20, I read dosMbhdvatvena for do- 4 Wrong knowledge has dosha- 

sMWidvasahakritatvena. IMva or the presence of a " defect " 

2 Anyathdsiddhatvam means ni- as its cause, in addition to the com- 
yatapurvavartitve sati andvasyakat- mon causes. 



ledge, it is not produced by any other cause (as " defect," 
&c.) The following is our argument as drawn out in 
full: Eight knowledge is not produced by any other 
instrumental causes than those of knowledge, while, at 
the same time, it is produced by these, because it is not 
the site of wrongness of knowledge, just like a jar. 1 ISTor 
can Udayana s 2 argument be brought forward as establish 
ing the dependence of authoritativeness on something 
external, for it is swallowed up by the dragon of the 
equally potent contradictory argument. " Eight know 
ledge is not produced by any cause which is other than 
the causes of knowledge and is also other than defect, 3 
from the very fact of its being knowledge like wrong 
knowledge." Again, since right knowledge can arise from 
the causes of knowledge per se, it would be a needless com 
plexity to suppose that anything else is a cause, whether 
you call it a guna or the absence of a " defect " (dosha)* 

" But surely if the presence of a defect is the cause of 
wrong knowledge, it is difficult to deny that its absence 
must be a cause of right knowledge ? " We meet this, 
however, by maintaining that the absence of defect is only 
an indirect and remote cause, as it only acts negatively by 
preventing wrong knowledge. As it has been said 

1 Wrongness of knowledge (apra- 2 I suppose this is the argument 

mdtva] can only reside in knowledge given at the close of the previous 

as a characteristic or quality thereof ; long purva-paksha. 

it cannot reside in a jar. The jar 3 These words " and is other than 

is, of course, produced by other in- defect " (dosha - vyatirikta) are, of 

strumental causes than those of course, meaningless as far as right 

knowledge (as, e.g., the potter s stick, knowledge is concerned; they are 

&c.), but it is not produced by these simply added to enable the author 

other causes in combination with to bring in " wrong knowledge" as 

being also produced by the instru- an example. Wrong knowledge is 

mental causes of knowledge (with caused by the causes of knowledge 

which it has nothing directly to do) ; plus " defect ; " right knowledge by 

and so by a quibble, which is less the former alone, 

obvious in Sanskrit than in English, 4 The Nya"ya holds that wrong 

this wretched sophism is allowed to knowledge is produced by a "defect," 

pass muster. The jar is not produced- as jaundice, &c., in the eye, and 

by - any- other - instrumental - causes- right knowledge by a gwia or " vir- 

than- those -of- knowledge, -while-at- tue" (as the direct contact of the 

the - same - time - it - is - produced - by- healthy organ with a true ob j ect) , or 

these. by the absence of a "defect." 


"Therefore we reasonably conclude from the presence 
of gunas the absence of defects, l from their absence 
the non-existence of the two kinds of non-authori- 
tativeness, 2 and from this the general conclusion." 3 

(&.) We maintain that the recognition of right know 
ledge is produced by the same causes only which make 
us perceive the first knowledge 4 [sc. the eye, mind, &c.] 
Nor can you object that this view is precluded, because it 
would imply that there could be no such thing as doubt ; 
for we answer that doubt arises in cases where, although 
all the causes which produce knowledge are present, there 
is also the simultaneous presence of some opposing cause, 
as a " defect," &c. 

As for your argument [0 Naiyayika ! given supra, in p. 
198, lines 17-24], I ask, Is your own argument an authori 
tative proof by itself or not ? If it is, it proves too much 
[for it would properly apply to itself and lead us to infer its 
own dependence on external proof, whereas you hold it to 
be independent of such] ; and if it is not, we should have a 
case of regressus in infinitum, for it will want some other 
proof to confirm its authoritativeness, and this too in its 
turn will want some fresh proof, and so on for ever. 

As for the argument urged by Udayana 5 in the Kusu- 
marijali, when he tries to establish that immediate and 
vehement action does not depend on the agent s certainty 
as to the authoritativeness of the speech which sets him 
acting : " Action depends on wish, its vehemence on that 

1 The guna (or jSeXrtVrT/ ets) of a jar, "the second knowledge is the 
an organ is not properly a cause of cognition of this perception in the 
pramd but rather doshdbhdva-bod- form " I perceive the jar ; " and 
haJca. simultaneously with it arises the 

2 Scil. "doubtful" (sandigdha) and cognition of the truth of the percep- 
" ascertained non-authoritativeness " tion, i.e., its authoritativeness or 
(nischitdprdmdnya). prdmdnya. 

3 Utsarga is a general conclusion 5 This seems to be a quotation of 
which is not necessarily true in every Udayana s own words, and no doubt 
particular case ; but here it means is taken from his very rare prose 
the con elusion that "right knowledge commentary on the Kusumdiijali, a 
has no special causes but the common specimen of which I printed in the 
causes of knowledge, the eye," &c. preface to my edition. This passage 

4 The first knowledge is " This is must come from the fifth book (v. 6 ?) 


of the wish, 1 wish on the knowledge that the thing wished 
for is a means to attain some wished-for end, and this is 
only ascertained by an inference based on some sign which 
proves that the thing is closely connected with the wished- 
for end, and this inference depends on the things being 
in direct contact with the agent s senses ; but throughout 
the whole series of antecedent steps the Mimamsa idea of 
the perception of authoritativeness is never once found as 
a cause of action." All this appears to us simple bluster, 
like that of the thief who ostentatiously throws open all 
his limbs before me, when I had actually found the gold 
under his armpit. It is only the knowledge that the thing 
is a means to attain the desired end, and this knowledge 
recognised as authoritative and right knowledge, which 
causes the definite volition to arise at all ; and in this we 
can distinctly trace the influence of that very perception 
of authoritativeness [whose existence he so vehemently 
pretended to deny]. If unhesitating action ever arose in 
any case from doubt, then, as it might always arise so in 
every given case, all ascertainment of authoritativeness 
would be useless ; and as the very existence of what is 
unascertained is rendered uncertain, poor authoritative- 
ness would have to be considered as dead and buried ! 
But enough of this prolix controversy ; since it has been 

" Therefore the authoritativeness of a cognition, which 
(authoritativeness) presented itself as representing 
a real fact, may be overthrown by the perception 
of a defect, which perception is produced by some 
sign that proves the discrepancy between the cog 
nition and the fact." 2 

Now with regard to the Yeda, which is the self-proved 
and authoritative criterion in regard to duty, [we have the 
following divergency between the two great Mimamsa 

1 I read tat-prdchuryam for tot- authoritativeness is self -proved, non- 
prdchurye in p. 1 34, line 7. authoritativeness is proved from 

2 This stanza affirms that accord- something else (as inference, &c.) 
ing to the Mimdmsa school, while 


schools] : The Veda is composed of three portions, respec 
tively called "hymns" (mantra), "explanatory passages" 
(arthavdda), and " injunctions " (vidhi) ; and by " injunc 
tion " we mean such sentences as " Let him who desires 
heaven sacrifice with the jyotishtoma." Here ta t the affix 
of the third person singular, denotes an enjoining power, 
which is " coloured " [or rendered definite] by the meaning 
of the root, according to the opinion of the followers of 
Bhatta Kumarila, who maintain that words signify 1 some 
thing definite by themselves [apart from the sentence]. 
The followers of Guru Prabhakara, on the contrary, hold 
that the whole sentence is a command relating to the 
sacrifice, as they maintain that words only signify an 
action or something to be done. 2 Thus all has been made 
plain. E. B. C. 

1 I take vyutpatti here as used for i.e., the bovine genus as connected 
inkti ; siddhe means yliatddau. with "bringing." We cannot have 

2 These are the two great Mim- a case of a noun without some 
a"msd, schools. The former, called governing verb, and vice versa. Cf. 
abhihitdnvai/a-vddinah, hold (like Waitz, as quoted by Professor Sayce 
the Naiydyika school) that words by (Comparative Philology, page 136): 
themselves can express their sepa- " We do not think in words but in 
rate meaning by the function abhidhd sentences ; hence we may assert 
or " denotation ; " these are subse- that a living language consists of 
quently combined into a sentence sentences, not of words. But a 
expressing one connected idea. The sentence is formed not of single 
latter, called an vitdbhidhdna-vddina h, independent words, but of words 
hold that words only express a mean- which refer to one another in a par 
ing as parts of a sentence and gram- ticular manner, like the correspond- 
matically connected with each other ; ing thought, which does not consist 
they only mean an action or some- of single independent ideas, but of 
thing connected with an action. In such as, connected, form a whole, and 
gam dnaya, gdm does not properly determine one another mutually." 
mean gotvci, but dnayandnvita-gotva, 

( 20 3 ) 



IF any one asks, " Where are we to learn how to separate 
a root and an affix so as to be able to say, This part is the 
original root and this is an affix/ " may we not reply that 
to those who have drunk the waters of Patanjali this 
question produces no confusion, since it is notorious that 
the rules of grammar have reference to this very point of 
the separation of the original roots and affixes ? Thus the 
very first sentence of the venerable Patanjali, the author 
of the " Great Commentary," is " atha sabddnusdsanam," 
" Now comes the exposition of words." The particle atha 
(" now ") is used here as implying a new topic or a com 
mencement ; and by the phrase, " exposition of words," is 
meant the system of grammar put forth by Panini. Now 
a doubt might here arise as to whether this phrase implies 
that the exposition of words is to be the main topic or 
not ; and it is to obviate any such doubt that he employed 
the particle atha, since this particle implies that what 
follows is to be treated as the main topic to the exclusion 
of everything else. 

The word " exposition " (anusdsana), as here used, im 
plies that thereby Vaidic^words, such as those in the line 
6am no devir alUshtaye? &c., and secular words as ancillary 
to these, as the common words for " cow," " horse," " man," 

1 Mddhava uses this peculiar term is eternal. He therefore treats of 

because the grammarians adopted spliota here, and not in his Jaimmi 

and fully developed the idea of the chapter. 
Purva-Mimdmssi school that sound 2 Rig-Veda, x. 9. 4. 


" elephant," " bird," &c., are made the subject of the exposi 
tion, i.e., are deduced from their original roots and properly 
formed, or, in other words, are explained as divided into 
root and affix. We must consider that the compound in 
this phrase represents a genitive of the object [sabddnusd- 
sanam standing for sabdasydnusdsanani], and as there is a 
rule of Panini (karmani cha, ii. 2, 14), which prohibits 
composition in such a construction, we are forced to con 
cede that the phrase sabddnusdsanam does not come before 
us as a duly authorised compound. 

Here, however, arises a discussion [as to the true appli 
cation of the alleged rule of Panini], for we hold that, by 
ii. 3, 66, wherever an object and an agent are both ex 
pressed in one and the same sentence in connection with 
a word ending with a krit affix, there the object alone can 
be put in the genitive and not the agent f 1 this limitation 
arising from our taking ubhayaprdpti in the sutra as a 
laliuvrihi compound?. Thus we must say, " Wonderful is 
the milking of cows by an unpractised cowherd." We 
may, however, remark in passing that some authors do 
maintain that the agent may in such cases be put in the 
genitive (as well as the object) ; hence we find it stated in 
the Kasika Commentary : " Some authors maintain that 
there should be an option in such cases without any dis 
tinction, and thus they would equally allow such a con 
struction as the exposition of words of the teacher or ly 
the teacher. " Inasmuch, however, as the words of the 
phrase in question really mean that the "exposition" 
intended relates to words and not to things, and since this 
can be at once understood without any mention of the 

Qffabddnusdsana, if judged by the we cannot say dsckaryo godoho siTcshi- 

apparent sense of Panini, ii. 2, 14, tena gopdlena (as it would violate ii. 

would be a wrong compound ; but 2, 14), neither can we say dscharyo 

it is not so, because ii. 2, 14 must be gavdm doJw sikshitasya gopdlasya (as 

interpreted in the sense of ii. 3, 66, it would violate ii. 3, 66). 

whence it follows that the compound (J* That is, the uWiayaprdpti of ii. 

would only be wrong if there were 3, 66, is a bahuvrihi agreeing with 

an agent expressed as well as an kriti in ii. 3, 65. These points are 

object, i.e., if such a word as dclidr- all discussed at some length in the 

yena followed. In the example given, Commentaries on Panini. 



agent, i.e., the teacher, any such mention would be plainly 
superfluous; and therefore as the object and the agent 
are not both expressed in one and the same sentence, this 
is not an instance of the genitive of the object (coming 
under ii. 3, 66, and ii. 2, 14), but rather an instance of 
quite another rule, viz., ii. 3, 65, which directs that an 
agent or an object, in connection with a word ending with 
a krit affix, is to be put in the genitive [which in this 
instance is expressed by the tatpuruska compound] ; and 
the compound in question will be strictly analogous to 
such recognised forms as idJima-pravraschana, paldsa-sd- 
tana, && Or we might argue that the genitive case 
implied in this sliashtliitatpurusha is one of the class 
called "residual," in accordance with Panini s rule (ii. 3, 
50), " Let the genitive be used in the residuum," [i.e., in 
the other constructions not provided for by special rules]@ 
and in this way we might defend the phrase against the 
opponent s attack. "But," it might be replied, "your 
alleged residual genitive could be assumed everywhere, 
and we should thus find all the prohibitions of composi 
tion in constructions with a genitive case rendered utterly 
nugatory." This we readily grant, and hence Bhartrihari 
in his Vakyapadtya has shown that these rules are mainly 
useful where the question relates to the accent 3 To this 
effect are the words of the great doctor Vardhaniana 

" In secular utterances men may proceed as they will, 

" But in Vaidic paths let minute accuracy of speech be 

"Thus have they explained the meaning of Panini s 
sutras, since 

" He himself uses such phrases as janikartuh and tat- 

\ * These actually occur in the Com- ( 4 /These compounds occur in Pa"- 

mentaries to Panini, ii. 2, S ; iii. 3, nini s own sutras (i. 4, 30, and i. 4, 

117, &c. 55), and. would violate his own rule 

2 This takes in all cases of rela- in ii. 2, 15, if we were to interpret 
tion, sambandha (i.e., shashthi-sam- the latter without some such saving 
bandha). modification as shashthi seshe. 

3 As in such rules as vi. 2, 139. 


Hence it follows that the full meaning of the sentence 
in question (of the MahabTidsJiya) is that "it is to be 
understood that the rules of grammar which may be 
taken as a synonym for the exposition concerning words 
are now commenced." 

"Well, then, for the sake of directly understanding 
this intended meaning, it would have been better to have 
said now comes grammar, as the words now comes 
the exposition of words involve a useless excess of 
letters." This objection cannot, however, be allowed, since 
the employment of such a word as sabddnusdsanam, 
the sense of which can be so readily inferred from its 
etymology, proves that the author intends to imply an 
end which shall establish that grammar is a subordinate 
9 study (anga) to the Veda Otherwise, if there were no 
such end set forth, there would be no consequent applica 
tion of the readers to the study of grammar. Nor may 
you say that this application will be sufficiently enforced 
by the injunction for study, " the Veda with its six sub 
ordinate parts must be read as a duty without any (special) 
end, because, even though there be such an injunction, 
it will not follow that students will apply to this study, if 
no end is mentioned which will establish that it is an 
anga of the Veda. /^Thus in old times the students, after 
reading the Veda, usScl to be in haste to say 

"Are not Vaidic words established by the Veda and 
secular by common life, 

" And therefore grammar is useless ? " 

Therefore it was only when they understood it to be an 
anga of the Veda that they applied themselves to its 
study. So in the same way the students of the present 
day would not be likely to apply themselves to it either. 
It is to obviate this danger that it becomes necessary to set 
forth some end which shall, at the same time, establish 

(3) The very word sabda in sabdd- v.5- Compare Max Muller, Sansk. \Q 

nusdsanam implies the Veda, since Liter., p. 113. It is quoted as from 
this is pre-eminently sabda. the Veda in the Mahabhashya. 


that grammar is an aiiga of the Veda. ) If, when the end 
is explained, they should still not apply themselves, then, 
being destitute of all knowledge of the true formation of 
secular words, they would become involved in sin in the 
course of sacrificial acts, and would consequently lose their 
religious merit. Hence the followers of sacrifice read, " One 
who keeps up a sacrificial fire, on using an incorrect word, 
should offer an expiatory offering to Saraswati." Now it 
is to declare this end which establishes that it is an anga 
of the Veda that he uses the words atha sabddnusdsanam 
and not atha vydkaranam. Now the rules of grammar 
must have an end, and a thing s end is determined by men s 
pursuit of it with a view thereto. Just as in a sacrifice 
undertaken with a view to heaven, heaven is the end; in the 
same way the end of the exposition of words is instruction 
concerning words, i.e., propriety of speech. "But," an objec 
tor may say, "will not the desired end be still un attained 
for want of the true means to it ? JSTor can it be said 
that reading the Veda word by word is the true means ; 
for this cannot be a means for the understanding of words, 
since their number is infinite, as divided into proper and 
improper wordsC} Thus there is a tradition that Brihas- 
pati for a thousand divine years taught to Indra the study 
of words as used in their individual forms when the Veda 
is read word by woroX? : and still he came not to the end 
Here the teacher was Brihaspati, the pupil was Indra, and 
the time of study a thousand years of the gods ; and yet 
the termination was not reached, how much less, then, 
in our day, let a -man live ever so long? Learning is 
rendered efficient by four appropriate means, reading, 
understanding, practising, and handing it on to others ; 
but in the proposed way life would only suffice for the bare 
time of reading; therefore the reading word by word is 
not a means for the knowledge of words, and consequently, 

1 In the Calcutta text, p. 138, dele danda in line 3 after lhavet, and 
insert it in line 4 after sabddnain. 

A* in the BO-callc J ]_>ada text. 



as we said at first, the desired end is not established." 
We reply, however, that it was never conceded that the 
knowledge of words was to be attained by this reading 
word by word. And again, since general and special rules 
apply at once to many examples, when these are divided 
into the artificial parts called roots, &c. (just as one cloud 
rains over many spots of ground), in this way we can 
easily comprehend an exposition of many words. Thus, 
for instance, by the general rule (iii. 2, i), Jcarmani, the 
affix an is enjoined after a root when the object is in 
composition with it; and by this rule we learn many 
words, as JcumlfiaTcdra, u a potter," kdndaldva, " a cutter of 
stems," &c. But the supplementary special rule (iii. 2, 3), 
dto nupasarge kah, directing that the affix ka is to be used 
after a root that ends in long d when there is no upasarga, 
shows how impracticable this reading word by word would 
be [since it would never teach us how to distinguish an 
upasarga]. " But since there are other angas, why do you 
single out grammar as the one object of honour ? " We 
reply, that among the six angas the principal one is 
grammar, and labour devoted to what is the principal is 
sure to bear fruit. Thus it has been said 

" Nigh unto Brahman himself, the highest of all religious 

" The wise have called grammar the first anga of the 

Hence we conclude that the exposition of words is the 
direct end of the rules of grammar, but its indirect end is 
the preservation, &c., of the Veda. Hence it has been 
said by the worshipful author of the great Commentary 
[quoting a Varttika], " the end (or motive) is preservation, 
inference, scripture, facility, and assurance.",: 1 Moreover 
prosperity arises from the employment of a correct word ; 
thus Katyayana has said, "There is prosperity in the 
employment of a word according to the sdstra ; it is equal 
to the words of the Veda itself." Others also have said 

09 See Ballantyne s MalidUidsliya, pp. 12, 64. 


that "a single word thoroughly understood and rightly 
used becomes in Swarga the desire-milking cow/ Thus 
(they say) 

" They proceed to heaven, with every desired happiness, 
in well-yoked chariots of harnessed speech ; 

"But those who use such false forms as achikramata 
must trudge thither on foot/ i; 

ISTor need you ask " how can an irrational word possess 
such power ? " since we have revelation declaring that it 
is like to the great god. For the Sruti says, " Four are its 
horns, three its feet, two its heads, and seven its hands, 
roars loudly the threefold-bound bull, the great god enters 
mortals" (Eig-Veda, iv. 58, 3). The great commentator 
thus explains it : The " four horns " are the four kinds 
of words nouns, verbs, prepositions, and particles; its 
"three feet " mean the three times, past, present, and future, 
expressed by the tense-affixes, lat, &c. ; the " two heads," 
the eternal and temporary (or produced) words, distin 
guished as the "manifested" and the " manif ester ;" its 
" seven hands " are the seven case affixes, including the 
conjugational terminations; " threefold bound," as enclosed 
in the three organs the chest, the throat, and the head. 
The metaphor "bull" (vrishabha) is applied from its pouring 
forth (varshana), i.e., from its giving fruit when used with 
knowledge. "Loudly roars," i.e., utters sound, for the root 
ru means " sound ; " here by the word " sound " developed 
/" speech (or language).. 2 -^ implied; "the great god enters 
mortals," the "great god," i.e., speech, enters mortals, 
i.e., men endowed with the attribute of mortality. Thus is 
/ declared the likeness [of speech^ 3 to the supreme Brahman. 

The eternal word, called sphota, without parts, and the 
cause of the world, is verily Brahman ; thus it has been 

| A (j)AchiJcramata seems put here as Bhartrihari which immediately fol- 

a purposely false form of the fre- low. 

quentative of Icram for achankra- ( 3 One would naturally supply xal- 

dasya after sdmyam, but the Malui- 

Or it may mean " the developed bhdshya has nah sdmyam (see Bal- 
universe." Compare the lines of lantyne s ed., p. 27). 


declared by Bhartrihari in the part of his book called the 

" Brahman, without beginning or end, the indestructible 

essence of speech, 
" Which is developed in the form of things, and whence 

springs the creation of the world." 
" But since there is a well-known twofold division of 
words into nouns and verbs, how comes this fourfold 
division ? " We reply, because this, too, is well known. 
Thus it has been said in the Prakirnaka 

" Some make a twofold division of words, some a four 
fold or a fivefold, 
"Drawing them up from the sentences as root, affix, 

and the like." 

Helaraja interprets the fivefold division as including 
Jearmapmvachaniya&D But the fourfold division, men 
tioned by the great commentator, is proper, since karma- 
pravachanfyas distinguish a connection produced by a 
particular kind of verb, and thus, as marking out a par 
ticular kind of connection and so marking out a particular 
kind of verb, they are really included in compounded 
prepositions (upasargasfe) 

" But," say some, " why do you talk so much of an 
eternal sound called sphota? This we do not concede, 
since there is no proof that there is such a thing." We 
reply that our own perception is the proof. Thus there 
is one word " cow," since all men have the cognition of a 
word distinct from the various letters composing it. You 
cannot say, in the absence of any manifest contradiction, 
that this perception of the word is a false perception. 

( l I.e., prepositions used separately ample, S dkalyasamhitdm anu prd- 

as governing cases of their own, and varshat, "he rained after the Sakalya 

not (as usually in Sanskrit) in com- hymns," anu implies an understood 

position. ver b nisamya, "having heard, and 

v 2 The karmaprarachanlyas imply this verb shows that there is a rela- 

aTverb other than the one expressed, tion of cause and effect between the 

and they are said to determine the hymns and the rain. This anu is 

relation which is produced by this said to determine this relation, 
understood verb. Thus in the ex- 


Hence you must concede that there is such a thing as 
sphota, as otherwise you cannot account for the cognition 
of the meaning of the word. For the answer that its 
cognition arises from the letters cannot bear examination, 
since it breaks down before either horn of the following 
dilemma : Are the letters supposed to produce this cog 
nition of the meaning in their united or their individual 
capacity ? Not the first, for the letters singly exist only 
for a moment, and therefore cannot form a united whole 
at all ; and not the second, since the single letters have no 
power to produce the cognition of the meaning [which the 
word is to convey]. There is no conceivable alternative 
other than their single or united capacity ; and therefore 
it follows (say the wise in these matters) that, as the 
letters cannot cause the cognition of the meaning, there 
must be a sphota by means of which arises the knowledge 
of the meaning ; and this sphota is an eternal sound, dis 
tinct from the letters and revealed by them, which causes 
the cognition of the meaning. " It is disclosed (sphutyate) 
or revealed by the letters," hence it is called sphota, as 
revealed by the letters ; or " from it is disclosed the 
meaning," hence it is called sphota as causing the knowledge 
of the meaning, these are the two etymologies to explain 
the meaning of the word. And thus it hath been said by 
the worshipful Patanjali in the great Commentary, " Now 
what is the word cow gauh ? It is that word by which, 
when pronounced, there is produced the simultaneous 
cognition of dewlap, tail, hump, hoofs, and horns." This 
is expounded by Kaiyata in the passage commencing, 
" Grammarians maintain that it is the word, as distinct 
from the letters, which expresses the meaning, since, if 
the letters expressed it, there would be no use in pro 
nouncing the second and following ones [as the first would 
have already conveyed all we wished]," and ending, " The 
Vdkyapadiya has established at length that it is the sphota 
which, distinct from the letters and revealed by the sound, 
expresses the meaning. ^25 

Jr See Ballantyne s ed., p. 10. 


Here, however, an objector may urge, " But should we 
not rather say that the spJwta has no power to convey the 
meaning, as it fails under either of the following alterna 
tives, for is it supposed to convey the meaning when itself 
manifested or unmanifested ? Not the latter, because it 
would then follow that we should find the effect of con 
veying the meaning always produced, since, as spJwta is 
supposed to be eternal, and there would thus be an ever- 
present cause independent of all subsidiary aids, the effect 
could not possibly fail to appear. Therefore, to avoid this 
fault, we must allow the other alternative, viz., that spJwta 
conveys the meaning when it is itself manifested. Well, 
then, do the manifesting letters exercise this manifesting 
power separately or combined ? Whichever alternative 
you adopt, the very same faults which you alleged against 
the hypothesis of the letters expressing the meaning, will 
have to be met in your hypothesis that they have this 
power to manifest sphota. This has been said by Bhatta 
in his Mimamsa-sloka-varttika 

" The grammarian who holds that spJwta is manifested 
by the letters as they are severally apprehended, 
though itself one and indivisible, does not thereby 
escape from a single difficulty." 

The truth is, that, as Panini (i. 4, 14) and Gotama (Sut. 
ii. 123) both lay it down that letters only then form a 
word when they have an affix at the end, it is the letters 
which convey the word s meaning through the apprehen 
sion of the conventional association of ideas which they 
c^jQ hell If you object that as there are the same letters in 
rasa as in sara, in nava as in vana, in dind as in nadi, in 
mdra as in rdma, in rdja as in jdra, &c., these several 
pairs of words would not convey a different meaning, we 
reply that the difference in the order of the letters will 
produce a difference in the meaning. This has been said 
by Tautatita 

is not very clear, the anu and so imply the successive order of 
in anugraha might mean kramena, the letters. 


" As are the letters in number and kind, whose power 
is perceived in conveying any given meaning of 
a word, so will be the meaning which they 

Therefore, as there is a well-known rule that when the 
same fault attaches to both sides of an argument it cannot 
be urged against one alone, we maintain that the hypothesis 
of the existence of a separate thing called sphota is un 
necessary, as we have proved that it is the letters which 
express the word s meaning [your arguments against our 
view having been shown to be irrelevant]." 

All this long oration is really only like a drowning man s 
catching at a straw p for either of the alternatives is im 
possible, whether you hold that it is the single letters or 
their aggregation which conveys the meaning of the word. 
It cannot be the former, because a collection of separate 
letters, without any one pervading cause^ 2 could never 
produce the idea of a word any more than a collection of 
separate flowers would form a garland without a string. 
Nor can it be the latter, because the letters, being sepa 
rately pronounced and done with, cannot combine into 
an aggregate. For we use the term " aggregate " where a 
number of objects are perceived to be united together in 
one place ; thus we apply it to a Grislea tomentosa, an 
Acacia catechu, a Butea frondosa, &c., or to an elephant, 
a man, a horse, &c., seen together in one place ; but these 
letters are not perceived thus united together, as they are 
severally produced and pass away; and even on the 
hypothesis of their having a "manifesting" power, they 
can have no power to form an aggregate, as they can only 
manifest a meaning successively and not simultaneously. 
Nor can you imagine an artificial aggregate in the letters, 
because this would involve a " mutual dependence " (or 
reasoning in a circle); for, on the one hand, the letters 
would only become a word when their power to convey 

v>In the Calcutta edition, p. 142, (jt In P- 142, line 3, I add vind 
line 1 1, 1 read Jcalpam for kalpanam. after nimittam. 


one meaning had been established ; and, on the other hand, 
their power to convey one meaning would only follow 
when the fact of their being a word was settled. Therefore, 
since it is impossible that letters should express the mean 
ing, we must accept the hypothesis of sphota. " But even 
on your own hypothesis that there is a certain thing called 
sphota which expresses the meaning, the same untenable 
alternative will recur which we discussed before; and 
therefore it will only be a case of the proverb that the 
dawn finds the smuggler with the revenue-officer s house 
close by. Cv This, however, is only the inflation of the 
world of fancy from the wide difference between the two 
cases. For the first letter, in its manifesting power, 
reveals the invisible sphota, and each successive letter 
makes this sphota more and more manifest, just as the 
Veda, after one reading, is not retained, but is made sure 
by repetition ; or as the real nature of a jewel is not 
clearly seen at the first glance, but is definitely mani 
fested at the final examination. This is in accordance 
with the authoritative saying (of the teacher) : " The seed 
is implanted by the sounds, and, when. the idea is ripened 
by the successive repetition, the word is finally ascertained 
simultaneously with the last uttered letter." Therefore, 
since Bhartrihari has shown in his first book that the 
letters of a word [being many and successive] cannot 
manifest the meaning of the word, as is implied by the 
very phrase, "We gain such and such a meaning from 
such and such a word" we are forced to assume the exist- 
fj (f enceJobf an indivisible spliota as a distinct category, which 
has the power to manifest the word s meaning. All this 
has been established in the discussion (in the Mahabhashya) 
on " genus " (jdti), which aims at proving that the mean 
ing of all words is ultimately that summum genus, i.e., that 

\J* The ghatta is the place where house just as day dawns and is thus 

dues and taxes are collected. Some caught. Hence the proverb means 

one anxious to evade payment is uddesydsiddhi. 

going by a private way by night, 2 In p. 143, line 13, 1 read sphota- 

but he arrives at the tax-collector s kalhdvam for sphotdbhdvam. 


existence whose^characteristic is perfect knowledge of the 
supreme reality i- (Brahman). 

" But if all words mean only that supreme existence, then 
all words will be synonyms, having all the same meaning ; 
and your grand logical ingenuity would produce an aston 
ishing result in demonstrating the uselessness of human 
language as laboriously using several words to no purpose 
at the same time ! Thus it has been said 

"The employment of synonymous terms at the same 
time is to be condemned; for they only express 
their meaning in turn and not by combina 

" Therefore this opinion of yours is really hardly worth 
the trouble of refuting." 

All this is only the ruminating of empty ether ; for 
just as the colourless crystal is affected by different objects 
which colour it as blue, red, yellow, &c., so, since the sum- 
mum genus, Brahman, is variously cognised through its 
connection with different things, as severally identified 
with each, we thus account for the use of the various con 
ventional words which arise from the different species, 2 as 
cow, &c., these being "existence" (the summum genus) as 
found in the individual cow, &c. To this purport we 
have the following authoritative testimony 

" Just as crystal, that colourless substance, when seve 
rally joined with blue, red, or yellow objects, is 
seen as possessing that colour." 

And so it has been said by Hari, " Existence [pure and 
simple] being divided, when found in cows, &c., by reason 
of its connection with different subjects, is called this or 
that species, and on it all words depend. This they call 
the meaning of the stem and of the root. This is exist 
ence, this the great soul ; and it is this which the affixed 
tva t tal, &c., express" (Panini v. I, 119). 

(jOCf. Ballantyne s Transl. of the individual (vyakti) ; the Nyaya holds 
]\aha~bhashya, pp. 9, 32. that a word means an individual as 

C 2 The Mimamsa holds that a word distinguished by such and such a 

means the genus (jdti) and not the genus (or species). 


" Existence " is that great summum genus which is found 
in cows, horses, &c., differentiated by the various subjects 
in which it resides; and the inferior species, "cow," 
"horse/ &c., are not really different from it; for the 
species " cow " and " horse " (gotva and asvatva) are not 
really new subjects, but each is " existence " as residing 
in the subject " cow " and " horse." Therefore all words, as 
expressing definite meanings, ultimately rest on that one 
summum genus existence, which is differentiated by the 
various subjects, cows, &c., in which it resides ; and hence 
"existence" is the meaning of the stem- word (prdtipadika). 
A " root " is sometimes defined as that which expresses 
<-/ bhdva; 1 now, as bhdva is "existence," the meaning of a 
root is really existence;. 2 . Others say that a root should be 
defined as that which expresses " action " (kriyd) ; but here 
again the" meaning of a root will really be " existence," 
since this "action" will be a genus, as it is declared to 
reside in many subjects, in accordance with the common 
definition of a genus, in the line 

" Others say that action (Icriyd) is a genus, residing in 
many individuals." 

So, too, if we accept Panini s definition (v. I, 119), "Let 
the affixes tva and tal come after a word [denoting any 
thing], when we speak of the nature (blidvd) thereof," it is 
clear from the very fact that abstract terms ending in tva 
or td [as asvatva and asvatd] are used in the sense of bhdva, 
that they do express " existence." " This is pure exist 
ence " from its being free from all coming into being or 
ceasing to be; it is eternal, since, as all phenomena are 
developments thereof, it is devoid of any limit in space, 
time, or substance: this existence is called "the great 
soul." Such is the meaning of Hari s two kdrikds quoted 
above. So, too, it is laid down in the discussion on sam- 
landlia [in Hari s verses] that the ultimate meaning of all 

>2jCf. Rig- Veda Pratis. xii. 5. monly received definitions of some 

v 2 He here is trying to show that grammatical terms, 
his view is confirmed by the com- 


words is that something whose characteristic is perfect 
knowledge of the real meaning of the word Substance. 

"The true Eeality is ascertained by its illusory forms; the 
true substance is declared by words through illusory dis- 
- guises; as the object, Devadatta s house/ is apprehended 
by a transitory cause of discrimination, 1 but by the word 
house itself, the pure idea [without owners] is expressed. 

So, too, the author of the Mahabhashya, when explaining 
the VarttiksClP" a word, its meaning, and its connection 
being fixed," in the passage beginning " substance is eter 
nal," has shown that the meaning of all words is Brahman, 
expressed by the word " substance " and determined by 
various unreaO-* conditions [as " the nature of horse," &c.] 

According to the opinion of Vajapyayana, who main- 
^ tains that all words mean a genus, words like " cow," 
1 : &jr denote a genus which resides by intimate relation in 
different substances ; and when this genus is apprehended, 
through its connection with it we apprehend the particular 
substance in which it resides. Words like " white," &c., 
denote a genus which similarly resides in qualities; through 
the connection with genus we apprehend the quality, and 
through the connection with the quality we apprehend 
the individual substance. So in the case of words express 
ing particular names, in consequence of the recognition 
that " this is the same person from his first coming into 
existence to his final destruction, in spite of the difference 
produced by the various states of childhood, youth, adoles 
cence, &c.," we must accept a fixed genus as Devadatta- 
f hood, 6 &c. [as directly denoted by them]. So, too, in words 
expressing " action " a genus is denoted ; this is the root- 
meaning, as in patkati, " he reads," &c., since we find here 
a meanin common to all who read. 

(iJSince Devadatta is only its 4 In p. 145, line 8, read asatya ^ 4 

transient owner. for asvattha. 

2 So by the words "horse," "cow," 5 We have here the well-known 
&c., Brahman is really meant, the four grammatical categories, jdti, 
one abiding existence. guna, dravya or sanjnd, and Tcriyd. 

3 Cf. Ballantyne s Mahdbh&hya, v 6 But cf. Siddh. Muktdv., p. 6, 
pp. 44, 50. line 12. 


In the doctrine of Yyadi, who maintained that words 
meant individual things [and not classes or genera], the 
individual thing is put forward as that which is primarily 
denoted, while the genus is implied [as a characteristic 
mark] ; and he thus avoids the alleged faults of " indefinite- 
ness," and " wandering away from its proper subject. ^ 1 

Both views are allowed by the great teacher Panini; 
since in i. 2, 58, he accepts the theory that a word means 
the genus, where he says that " when the singular is used 
to express the class the plural may be optionally used " 
[as in the sentence, " A Brahman is to be honoured," which 
may equally run, " Brahmans are to be honoured "] ; while 
in i. 2, 64, he accepts the theory that a word means the 
individual thing, where he says, " In any individual case 
there is but one retained of things similar in form " [i.e., 
the dual means Rama and Rama, and the plural means 
Rama, and Rama and Rama; but we retain only one, 
adding a dual or plural affix]. Grammar, in fact, being 
adapted to all assemblies, can accept both theories with 
out being compromised. Therefore both theories are in a 
sense true; 2 but the real fact is that all words ultimately 
mean the Supreme Brahman. 

As it has been said 

" Therefore under the divisions of the meanings of words, 
one true universal meaning, identical with the one 
existent, shines out in many forms as the thing 

Hari also, in his chapter discussing sambandha, thus 
describes the nature of this true meaning 

l Thus we read in the Siddhdnta should not include ; if it is held to 

Muktavali, p. 82, that the Mimdmsa mean many individuals, it will have 

holds that a word means the genus an endless variety of meanings and 

and not the individual, since other- be -" indefinite." 

wise there would be vyabhichdra and t^/This seems the meaning of the 

dnantya (cf. also Mahesachandra text as printed tasmdt dvayam sat- 

Nyayaratna s note, KaVya-praksa, yam, but I should prefer to read 

p. 10). If a word is held to mean conjecturally tasmdd advayam sat- 

only one individual, there will be the yam, "therefore non-duality is the 

first fault, as it will "wander away" truth." 
and equally express others which it 


"That meaning in which the subject, the object, and 
the perception [which unites them] are insuscep 
tible of doub^Ptffo^ only is called the truth by 
those who know the end of the three Yedas." 
So too in his description of substance, he says 
" That which remains as the Eeal during the presence 
of modification, as the gold remains under the 
form of the earring, that wherein change comes 
and goes, that they call the Supreme Nature." 
The essential unity of the word and its meaning is 
maintained in order to preserve inviolate the non-duality 
of all things which is a cardinal doctrine of our philo 

"This [Supreme Nature] is the thing denoted by all 
words, and it is identical with the word ; but the relation 
of the two, while they are thus ultimately identical, varies 
; as does the relation of the two souls. ! 2 , 

The meaning of this Karika is that Brahman is the 
one object denoted by all words ; and this one object has 
various differences imposed upon it according to each 
particular form; but the conventional variety of the 
differences produced by these illusory conditions is only 
the result of ignorance. Non-duality is the true state ; 
but through the power of " concealment ^ [exercised by 
illusion] at the time of the conventional use of words a 
manifold expansion takes place, just as is the case during 
sleep. Thus those skilled in Vedanta lore tell us 
" As all the extended world of dreams is only the 
development of illusion in me, so all this extended 
waking world is a development of illusion like 

When the unchangeable Supreme Brahman is thus 
known as the existent joy-thought and identical with the 
individual soul, and when primeval ignorance is abolished, 

V 1 Scil. they can only be the absolute 3 The Samvriti of the text seems 

Brahman who alone exists. to correspond to the dvarana so fre- 

2 Scil. the individual soul (jiva) quent in Veddnta books, 
and Brahman. 


final bliss is accomplished, which is best defined as the 
abiding in identity with this Brahman, according to the 
text, "He who is well versed in the Word-Brahman 
60attains to the Supreme Brahman.T 1 And thus we estab 
lish the fact that the " exposition of words " is the means 
to final bliss. 

Thus it has been said 

" They call it the door of emancipation, the medicine 
of the diseases of speeclythe purifier of all sciences, 
the science of sciences.? 2 
And so again 

" This is the first foot-round of the stages of the ladder 
of final bliss, this is the straight royal road of the 
travellers to emancipation." 

Therefore our final conclusion is that the Sastra of 
grammar should be studied as being the means for attain 
ing the chief end of man. E. B. C. 

v 1 *This passage is quoted in the Upanishad, i. 3, I, where it is ex- 
Maitri Upanishad, vi. 22. plained by ^amkara as vidydsv adhi 

(* Adhividyam occurs in Taitt. yad darsanam tad adhividyam. 



" BUT how can we accept the doctrine of illusory emana 
tion [thus held by the grammarians, following the guidance 
of the purva and uttara Mimamsa schools], when the 
system of development propounded by the Sankhyas is 
still alive to oppose it ? " Such is their loud vaunt. Now 
the Sastra of this school may be concisely said to maintain 
four several kinds of existences, viz., that which is evol 
vent l only, that which is evolute only, that -which is both 
evolute and evolvent, and that which is neither, (a.) Of 
these the first is that which is only evolvent, called the root- 
evolvent or the primary ; it is not itself the evolute of any 
thing else. It evolves, hence it is called the evolvent 
(prakriti) since it denotes in itself the equilibrium of the 
three qualities, goodness, activity, and darkness. This is 
expressed [in the Sankhya Karika], " the root-evolvent is 
no evolute." It is called the root-evolvent, as being both 
root and evolvent ; it is the root of all the various effects, 
as the so-called "great one," &c., but of it, as the primary, 
there is no root, as otherwise we should have a regressus 
ad infinitum. Nor can you reply that such a regressus ad 
infinitum is no objection, if, like the continued series of 
seed and shoot, it can be proved by the evidence of our 
senses, 2 because here there is no evidence to establish the 
hypothesis. (5.) The " evolutes and evolvents " are the 
great one, egoism, and the subtile elements, thus the 

1 I borrow this term from Dr. Hall. 

2 Compare Kusumanjali, L 4. 


Sankhya Karika ( 3), " the seven, the great one, &c., are 
evolute-evolvents." The seven are the seven principles, 
called the great one, &c. Among these the great prin 
ciple, called also the intellect, 1 &c., is itself the evolute of 
nature and the evolvent of egoism ; in the same manner 
the principle egoism, called also " self-consciousness " 
(abhimdna), is the evolute of the great one, intellect ; but 
this same principle, as affected by the quality of dark 
ness, is the evolvent of the five rudiments called subtile 
elements ; and, as affected by the quality of goodness, it 
is the evolvent of the eleven organs, viz., the five organs 
of perception, the eye, ear, nose, tongue, and skin; the five 
organs of action, the voice, hands, feet, anus, and genera 
tive organ ; and the mind, partaking of the character of 
both; nor can you object that in our arrangement the 
third quality, activity, is idle, as it acts as a cause by 
producing action in the others. This has been thus 
declared by f svara Krishna in his Karikiis 2 ( 24-27), 
" Self-consciousness is egoism. Thence proceeds a two 
fold creation, the elevenfold set and the five elemental 
rudiments. From modified 3 egoism originates the class of 
eleven imbued with goodness ; from egoism as the source 
of the elements originate the rudimentary elements, and 
these are affected by darkness ; but it is only from egoism 
as affected by activity that the one and the other rise. 
The intellectual organs are the eyes, the ears, the nose, the 
tongue, and the skin ; those of action are the voice, feet, 
hands, anus, and organ of generation. In this set is mind, 
which has the character of each; it determines, and it 
is an organ (like the other ten) from having a common 

1 One great defect in the Sdnkhya 2 It is singular that this is Mad- 
nomenclature is the ambiguity be- hava s principal Sjtnkhya authority, 
tween the terms for intellect (buddhi) and not the Sdnkhya Sutras, 
and those for mind (manas). Mad- 3 VaUcrita is here a technical term 
hava here applies to the former the meaning that goodness predominates 
term antahkarana or " internal over darkness and activity. On 
organ," the proper term for the this Kdrikd, comp. Dr. Hall s pre- 
latter. I have ventured to alter it face to the Sdukhya-sara, pp. 30- 
in the translation. 35. 


property with them." 1 All this has been explained at 
length by the teacher Vachaspati Misra in the Sankhya- 

(c.) The " evolute only " means the five gross elements, 
ether, &c., and the eleven organs, as said in the Karika, 
" The evolute consists of sixteen ; " that is, the set of six 
teen is evolute only, and not evolvent. Although it may 
be said that earth, &c., are the evolvents of such produc 
tions as cows, jars, &c., yet these are not a different " prin 
ciple" (tattva) from earth, &c., and therefore earth, &c., 
are not what we term " evolvents ; " as the accepted idea 
of an evolvent is that which is the material cause of a 
separate principle ; and in cows, jars, &c., there is the 
absence of being any such first principle, in consequence 
of their being all alike gross [i.e., possessed of dimensions] 
and perceptible to the senses. The five gross elements, 
ether, &c., are respectively produced from sound, touch, 
form, taste, and smell, each subtile element being accom 
panied by all those which precede it, and thus the gross 
elements will have respectively one, two, three, four, and 
five qualities. 2 The creation of the organs has been pre 
viously described. This is thus propounded in the San- 
khya Karika ( 22) 

" From nature springs the great one, from this egoism, 
from this the set of sixteen, and from five among 
the sixteen proceed the five gross elements." 

(d.) The soul is neither, as is said in the Karika, "The 
soul is neither evolvent nor evolute." That is, the soul, 
being absolute, eternal, and subject to no development, is 
itself neither the evolvent nor the evolute of aught beside. 
Three kinds of proof are accepted as establishing these 
twenty-five principles ; and thus the Karika ( 4). 

" Perception, inference, and the testimony of worthy 
persons are acknowledged to be the threefold proof, for 

1 As produced, like them, from 2 Cf. Colebrooke Essays, vol. i. p. 

modified egoism. The reading sam- 256. The tanmdtras will reproduce 

kalpavikalpdtmakam must be cor- themselves as the respective qualities 

rected by the Saukhya Kdrika". of the gross elements. 


they comprise every mode of demonstration. It is from 
proof that there results belief of that which is to be 

Here a fourfold discussion arises as to the true nature 
of cause and effect. The Saugatas 1 maintain that the 
existent is produced from the non-existent; the Naiya- 
yikas, &c., that the (as yet) non-existent is produced from 
the existent ; the Vedantins, that all effects are an illusory 
emanation from the existent and not themselves really 
existent; while the Sankhyas hold that the existent is 
produced from the existent. 

(a.) Now the first opinion is clearly untenable, since 
that which is itself non-existent and unsubstantial can 
never be a cause any more than the hare s horn; and, again, 
the real and unreal can never be identical. 

(6.) Nor can the non-existent be produced from the 
existent ; since it is impossible that that which, previous 
to the operation of the originating cause, was as non 
existent as a hare s horn should ever be produced, i.e., 
become connected with existence; for not even the cleverest 
man living can make blue yellow. 2 If you say, " But are 
not existence and non-existence attributes of the same 
jar ? " this is incorrect, since we cannot use such an 
expression as " its quality " in regard to a non-existent 
subject, for it would certainly imply that the subject 
itself did exist. Hence we conclude that the effect is 
existent even previously to the operation of the cause, 
which only produces the manifestation of this already 
existent thing, just like the manifestation of the oil in 
sesame seed by pressing, or of the milk in cows by milk 
ing. Again, there is no example whatever to prove the 
production of a thing previously non-existent. 

Moreover, the cause must produce its effect as being 
either connected with it or not connected ; in the former 

1 A name of the Buddhists. cannot be made a cow, nor a woman 

2 I.e., the nature of a thing (Sva- a man. 
Ihdva) cannot be altered a man 


alternative the effect s existence is settled by the rule 
that connection can only be between two existent things ; 
in the latter, any and every effect might arise from any 
and every cause, as there is nothing to determine the 
action of an unconnected thing. This has been thus put 
by the Sankhya teacher: " From the supposed non-exist 
ence of the effect, it can have no connection with causes 
which always accompany existence; and to him who 
holds the production of a non-connected thing there arises 
an utter want of determinateness." If you rejoin that " the 
cause, though not connected with its effect, can yet pro 
duce it, where it has a capacity of so doing, and this capa 
city of producing is to be inferred from seeing the effect 
actually produced," still this cannot be allowed, since in 
such a case as " there is a capacity for producing oil in 
sesame seeds," you cannot determine, while the oil is 
non-existent, that there is this capacity in the sesame 
seeds, whichever alternative you may accept as to their 
being connected or not with the oil [since our before-men 
tioned dilemma will equally apply here]. 

From our tenet that the cause and effect are identical, 
it follows that the effect does not exist distinct from the 
cause ; thus the cloth is not something distinct from the 
threads, as it abides in the latter [as its material cause] ; 
but where this identity is not found, there we do not find 
the relation of cause and effect ; thus a horse and a cow are 
distinct from each other [for one is not produced from the 
other, and therefore their qualities are not the same]; but 
the cloth is an acknowledged effect, and therefore not any 
thing different from its cause. 1 If you object that, if this 
were true, the separate threads ought to fulfil the office of 
clothing, we reply, that the office of clothing is fulfilled by 
the threads manifesting the nature of cloth when they are 
placed in a particular arrangement. As the limbs of a 
tortoise when they retire within its shell are concealed, 

i I take arthdntaram here as kavdchaspati s note, Tattva Kau- 
simply Ihinnam (cf. Tardnatha Tar- mudi, p. 47). 



and, when they come forth, are revealed, so the particular 
effects, as cloth, &c., of a cause, as threads, &c., when they 
come forth and are revealed, are said to be produced ; and 
when they retire and are concealed, they are said to be 
destroyed ; but there is no such thing as the production 
of the non-existent or the destruction of the existent. As 
has been said in the Bhagavad Gita (ii. 16) 

" There is no existence for the non-existent, nor non- 
existence for the existent." 

And, in fact, it is by inference from its effects that we 
establish the existence of the great evolvent, Nature (pra- 
kriti). This has been said [in the Karika, 9] 

" Effect exists, for what exists not can by no operation 
of cause be brought into existence ; materials, too, 
are selected which are fit for the purpose; every 
thing is not by every means possible ; what is 
capable does that to which it is competent; and 
like is produced from like." l 

Xor can we say [with the Vedantin] that the world is 
an illusory emanation from the one existent Brahman, 
because we have no contradictory evidence to preclude 
by its superior validity the primd facie belief that the 
external world is real [as we have in the case of mistaking 
a rope for a snake, where a closer inspection will discover 
the error] ; and again, where the subject and the attributed 
nature are so dissimilar as the pure intelligent Brahman 
and the unintelligent creation, we can no more allow the 
supposed attribution to be possible than in the case of 
gold and silver [which no one mistakes for each other]. 
Hence we conclude that an effect which is composed of 
happiness, misery, and stupidity, must imply a cause 
similarly composed ; and our argument is as follows : 
The subject of the argument, viz., the external world, must 
have a material cause composed of happiness, misery, and 
stupidity, because it is itself endued therewith ; whatever 
is endued with certain attributes must have a cause endued 

1 Colebrooke s translation. 


with the same, thus a ring has gold for its material cause, 
because it has the attributes of gold; our subject is a 
similar case, therefore we may draw a similar conclusion. 
What we call "being composed of happiness" in the 
external world is the quality of goodness; the "being 
composed of misery" is the quality of activity; 1 the 
"being composed of stupidity" is the quality of dark 
ness ; hence we establish our cause composed of the three 
qualities (i.e., prakriti, Nature). And we see that indi 
vidual objects are found by experience to have these three 
qualities; thus Maitra s happiness is found in his wife 
Satyavati, because the quality of "goodness" in her is 
manifested towards him; but she is the misery of her 
fellow- wives, because the quality of " activity " is mani 
fested towards them; while she causes indifference to 
Chaitra who does not possess her, because towards him 
the quality of "darkness" is manifested. So, too, in 
other cases also ; thus a jar, when obtained, causes us 
pleasure ; when seized by others it causes us pain ; but it 
is viewed with indifference by one who has no interest in 
it. Now this being regarded with no interest is what 
we mean by " stupidity," since the word moha is derived 
from the root muli, " to be confused," since no direct action 
of the mind arises towards those objects to which it is 
indifferent. Therefore we hold that all things, being 
composed of pleasure, pain, and stupidity, must have as 
their cause Nature, which consists of the three qualities. 
And so it is declared in the Svetasvatara Upanishad 
(iv. 5)- 

" The one unborn, for his enjoyment, approaches the 
one unborn (Nature) which is red, white, and black, 
and produces a manifold and similar offspring ; the 
other unborn abandons her when once she has been 

Here the words "red," "white," and "black," express 
the qualities " activity," " goodness," and " darkness," from 

1 Or "passion," rajas. 


their severally possessing the same attributes of colouring, 
manifesting, and concealing. 

Here, however, it may be objected, " But will not your 
unintelligent Nature, without the superintendence of some 
thing intelligent, fail to produce these effects, intellect, 
&c. ? therefore there must be some intelligent super 
intendent; and hence we must assume an all-seeing, 
supreme Lord." We reply that this does not follow, since 
even unintelligent Nature will act under the force of an 
impulse; and experience shows us that an unintelligent 
thing, without any intelligent superintendent, does act for 
the good of the soul, just as the unintelligent milk acts for 
the growth of the calf, or just as the unintelligent rain acts 
for the welfare of living creatures ; and so unintelligent 
Nature will act for the liberation of the soul. As it has 
been said in the Karika ( 57) 

" As the unintelligent milk acts for the nourishment of 
the calf, so Nature acts for the liberation of soul." 
But as for the doctrine of " a Supreme Being who acts 
from compassion," which has been proclaimed by beat of 
drum by the advocates of his existence, this has well-nigh 
passed away out of hearing, since the hypothesis fails to meet 
either of the two alternatives. For does he act thus before 
or after creation ? If you say " before," we reply that as 
pain cannot arise in the absence of bodies, &c., there will 
be no need, as long as there is no creation, for his desire to 
free living beings from pain [which is the main character 
istic of compassion] ; and if you adopt the second alterna 
tive, you will be reasoning in a circle, as on the one hand 
you will hold that God created the world through com 
passion [as this is His motive in acting at all], and on 
the other hand 1 that He compassionated after He had 
created. Therefore we hold that the development of 
unintelligent Nature [even without any intelligent super- 

i In other words on the one on the other hand it was the exist- 

hand the existing misery of beings ence of a created world which caused 

induced God to create a world in their misery at all. 
order to relieve their misery, and 


intendent] in the order of the series intellect, self-con 
sciousness, &c., is caused by the union of Nature and 
Soul, and the moving impulse is the good of Soul. Just 
as there takes place a movement in the iron in the prox 
imity of the unmoved magnet, so there takes place a 
movement in Nature in the proximity of the unmoved 
Soul ; and this union of Nature and Soul is caused by 
mutual dependence, like the union of the lame man and 
the blind man. Nature, as the thing to be experienced, 
depends on Soul the experiencer ; and Soul looks to final 
bliss, as it seeks to throw off the three kinds of pain, 
which, though really apart from it, have fallen upon it by 
its coining under the shadow of intellect through not 
recognising its own distinction therefrom. 1 This final 
bliss [or absolute isolation] is produced by the discrimina 
tion of Nature and Soul, nor is this end possible without it; 
therefore Soul depends on Nature for its final bliss. Just as 
a lame man and a blind man, 2 travelling along with a cara 
van, by some accident having become separated from 
their companions, wandered slowly about in great dismay, 
till by good luck they met .each other, and then the lame 
man mounted on the blind man s back, and the blind 
man, following the path indicated by the lame man, 
reached his desired goal, as did the lame man also, mounted 
on the other s shoulders; so, too, creation is effected by 
Nature and the soul, which are likewise mutually de 
pendent. This has been said in the Karika ( 21) 
"For the soul s contemplation of Nature and for its 
final separation the union of both takes place, as 
of the lame man and the blind man. By that 
union a creation is formed." 

" Well, I grant that Nature s activity may take place 
for the good of the soul, but how do you account for its 

1 Bondage, &c., reside in the in- piece of folk-lore. It is found in 

tellect, and are only rejected upon the Babylonian Talmud, Sanhedrim, 

soul through its proximity (cf. San- fol. 91, 6, and in the Gesta Roman- 

kkyapravachanabfidshya, i. 58). orum. 

~ This apologue is a widely spread 


ceasing to act ? " I reply, that as a wilful woman whose 
faults have once been seen by her husband does not return 
to him, or as an actress, having performed her part, retires 
from the stage, so too does Nature desist. Thus it is said 
in the Karika ( 59) 

"As an actress, having exhibited herself to the spec 
tators, desists from the dance, so does Nature desist, 
having manifested herself to Soul." 

For this end has the doctrine of those who follow 
Kapila, the founder of the atheistic Sankhya School, been 
propounded. E. B. C. 



WE now set forth, the doctrine of that school which pro 
fesses the opinions of such Munis as Patanjali and others, 
who originated the system of the Theistic Sankhya philo 
sophy. This school follows the so-called Yoga Sastra 
promulgated by Patanjali, and consisting of four chapters, 
which also bears the name of the "Sankhya Pravachana," or 
detailed explanation of the Sankhya. 1 In the first chapter 
thereof the venerable Patanjali, having in the opening 
aphorism, "Now is the exposition of Concentration" 
(yoga), avowed his commencement of the Yoga Sastra, 
proceeds in the second aphorism to give a definition of 
his subject, " Concentration is the hindering of the modi 
fications of the thinking principle," and then he expounds 
at length the nature of Meditation (samddhi). In the 
second chapter, in the series of aphorisms commencing, 
" The practical part of Concentration is mortification, 
muttering, and resignation to the Supreme," he expounds 
the practical part of yoga proper to him whose mind is not 
yet thoroughly abstracted (iii. 9), viz., the five external sub- 
servients or means, " forbearance," and the rest. In the 
third chapter, in the series commencing " Attention is the 
fastening [of the mind] on some spot," he expounds the 
three internal subservients attention, contemplation, and 
meditation, collectively called by the name " subjugation " 
(samyama), and also the various superhuman powers which 

i On this see Dr. Hall s Pref. to Sankhya Pr. Bhash., p. 20 ; S. Sara, p. 11. 


are their subordinate fruit. In the fourth chapter, in the 
series commencing, " Perfections spring from birth, plants, 
spells, mortification, and meditation," he expounds the 
highest end, Emancipation, together with a detailed account 
of the five so-called " perfections " (siddhis). This school 
accepts the old twenty-five principles [of the Sankhya], 
" Nature," &c. ; only adding the Supreme Being as the 
twenty-sixth a Soul untouched by affliction, action, fruit, 
or stock of desert, who of His own will assumed a body 
in order to create, and originated all secular or Vaidic 
traditions, 1 and is gracious towards those living beings who 
are burned in the charcoal of mundane existence. 

" But how can such an essence as soul, undefiled as the 
[glossy] leaf of a lotus, be said to be burned, that we should 
need to accept any Supreme Being as gracious to it ? " 
To this we reply, that the quality Goodness develops itself 
as the understanding, and it is this which is, as it were, 
burned by the quality Activity; and the soul, by the 
influence of Darkness, blindly identifying itself with this 
suffering quality, is also said itself to suffer. Thus the 
teachers have declared 

"It is Goodness which suffers under the form of the 

understanding and the substances belonging to 

Activity which torment, 2 
And it is through the modification of Darkness, as 

wrongly identifying, that the Soul is spoken of as 


It has been also said by Patanjali, 3 " The power of the 
enjoyer, which is itself incapable of development or of 
transference, in an object which is developed and trans 
ferred experiences the modifications thereof." 

Now the "power of the enjoyer" is the power of intel 
ligence, and this is the soul ; and in an object which is 

i i.e., he revealed the Veda, and 2 I read ye for te with Dr. Hall s 

also originated the meanings of MS. Tapya means rather " suscep- 

words, as well as instructed the tible of suffering." 

first fathers of mankind in the arts 3 This is really Vyjisa s comm. 

of life. on Sut., iv. 21. 


" developed" and "transferred," or reflected, i.e., in the 
thinking principle or the understanding, it experiences 
the modifications thereof, i.e., the power of intelligence, 
being reflected in the understanding, receives itself the 
shadow of the understanding, and imitates the modifica 
tions of it. Thus the soul, though in itself pure, sees 
according to the idea produced by the understanding ; and, 
while thus seeing at secondhand, though really it is dif 
ferent from the understanding, it appears identical there 
with. It is while the soul is thus suffering, that, by the 
practice of the eight subservient means, forbearance, reli 
gious observance, &c., earnestly, uninterruptedly, and for a 
long period, and by continued resignation to the Supreme 
Being, at length there is produced an unclouded recogni 
tion of the distinction between the quality Goodness and 
the Soul; and the five "afflictions," ignorance, &c., are 
radically destroyed, and the various "stocks of desert," 
fortunate or unfortunate, are utterly abolished, and, the 
undefiled soul abiding emancipated, perfect Emancipation 
is accomplished. 

The words of the first aphorism, " Now is the exposition 
of concentration," establish the four preliminaries which 
lead to the intelligent reader s carrying the doctrine into 
practice, viz., the object-matter, the end proposed, the 
connection [between the treatise and the object], and the 
person properly qualified to study it. The word " now" 
(atha) is accepted as having here an inceptive meaning, 
[as intimating that a distinct topic is now commenced]. 
" But," it may be objected, " there are several pos 
sible significations of this word atha ; why, then, should 
you show an unwarranted partiality for this particular 
inceptive meaning ? The great Canon for nouns and 
their gender [the Amara Kosha Dictionary] gives many 
such meanings. Atha is used in the sense of an auspi 
cious particle, after, now (inceptive), what? (interro 
gatively), and all (comprehensively). Now we willingly 
surrender such senses as interrogation or comprehensive- 


ness ; but since there are four senses certainly suitable, 
i.e., after, an auspicious particle, reference to a pre 
vious topic/ and the inceptive now, there is no reason 
for singling out the last." This objection, however, will not 
stand, for it cannot bear the following alternative. If you 
maintain the sense of " after," then do you hold that it 
implies following after anything whatever, or only after 
some definite cause as comprehended under the general 
definition of causation, 1 i.e., " previous existence [relatively 
to the effect] " ? It cannot be the former, for, in accord 
ance with the proverb that " No one stands for a single 
moment inactive," everybody must always do everything 
after previously doing something else ; and since this is at 
once understood without any direct mention at all, there 
could be no use in employing the particle atha to convey 
this meaning. Nor can it be the latter alternative ; be 
cause, although we fully grant that the practice of concen 
tration does in point of fact follow after previous tranquil 
lity, &c., yet these are rather the necessary preliminaries 
to the work of exposition, and consequently cannot have 
that avowed predominance [which the presumed cause 
should have]. " But why should we not hold that the 
word atha implies that this very exposition is avowedly 
the predominant object, and does follow after previous 
tranquillity of mind, &c. ? " We reply, that the aphorism 
uses the term. " exposition " (anusdsana), and this word, 
etymologically analysed, implies that by which the yoga 
is explained, accompanied with definitions, divisions, and 
detailed means and results ; and there is no rule that such 
an exposition must follow previous tranquillity of mind, 
&c., the rule rather being that, as far as the teacher is 
concerned, it must follow a profound knowledge of the 
truth and a desire to impart it to others ; for it is rather 
the student s desire to know and his derived knowledge, 
which should have quiet of mind, &c., as their precur 
sors, in accordance with the words of Sruti : " Therefore 

1 Cf. BJidshd-parichchheda, 15, a. 


having become tranquil, self-subdued, loftily indifferent, 
patient, full of faith and intent, let him see the soul in 
the soul." 1 Nor can the word atha imply the necessary 
precedence, in the teacher, of a profound knowledge of the 
truth and a desire to impart it to others ; because, even 
granting that both these are present, they need not to be 
mentioned thus prominently, as they are powerless in 
themselves to produce the necessary intelligence and effort 
in the student. Still [however we may settle these points] 
the question arises, Is the exposition of the yoga ascertained 
to be a cause of final beatitude or not ? If it is, then it is 
still a desirable object, even if certain presupposed condi 
tions should be absent ; and if it is not, then it must be un 
desirable, whatever conditions may be present. 2 But it is 
clear that the exposition in question is such a cause, since 
we have such a passage of the Sruti as that [in the Katha 
Upanishad, ii. 12]: " By the acquirement of yoga or in 
tense concentration on the Supreme Soul, the wise man 
having meditated leaves behind joy and sorrow;" and 
again, such a passage of the Smriti as that [in the Bhaga- 
vad Gita, ii. 5 3] : " The intellect unwavering in contem 
plation will then attain yoga." Hence we conclude that it 
is untenable to interpret atha as implying that the expo 
sition must follow " after " a previous inquiry on the part 
of the student, or " after " a previous course of ascetic 
training and use of elixirs, &c. [to render the body 

But in the case of the Yedanta Sutras, which open with 
the aphorism, " Now, therefore, there is the wish to know 
Brahman," Sankara Acharya has declared that the incep 
tive meaning of atha must be left out of the question, as 
the wish to know Brahman is not to be undertaken [at 
will] ; and therefore it must be there interpreted to mean 
" after," i.e., that this desire must follow a previous 

1 Satapatha Br., xiv. 7, 2, 28. different conditions which atha is 

2 I read in the second clause tad- supposed to assume as being neces- 
Ihdve pi, understanding by tad the sarily present. 


course of tranquillity, &c., as laid down by the well-known 
rule which enjoins the practice of tranquillity, self-control, 
indifference, endurance, contemplation, and faith, the object 
being to communicate the teaching to a proper student 
as distinguished by the possession of the four so-called 
" means." l 

" Well, then, let us grant that atha cannot mean c after; 
but why should it not be simply an auspicious particle ? " 
But this it cannot be, from the absence of any connection 
between the context and such auspicious meaning. Aus- 
piciousness implies the obtaining of an unimpeached and 
desired good, and what is desired is so desired as being the 
attainment of pleasure or the avoidance of pain ; but this 
auspiciousness cannot belong to the exposition of yoga, 
since it is in itself neither pleasure nor the cessation of 
pain. 2 Therefore it cannot be at all established that the 
meaning of the aphorism is that " the exposition of the 
yoga is auspicious ; " for auspiciousness cannot be either 
the primary meaning of atha or its secondary meaning by 
metonymy, since it is its very sound which is in itself 
auspicious [without any reference to the meaning], like 
that of a drum. " But why not say that just as an im 
plied meaning may enter into the direct meaning of a 
sentence, so an effect [like this of auspiciousness] may 
also be included, since both are equally unexpressed so far 
as the actual words are concerned ? " 3 We reply, that in 
the meaning of a sentence the connection must be between 
the meaning of one word and that of another ; otherwise 
we should be guilty of breaking the seal which the rule of 
the grammarians has set, that " verbal expectancy 4 can be 
fulfilled by ivords alone." 

1 These are, i. , the discrimination 3 Granting that atlia does not 
of the eternal from the phenomenal ; here mean " auspicious," why should 
ii., the rejection of the fruit of ac- not this be the implied meaning, 
tions here or hereafter ; iii., the pos- as all allow that the particle atha 
session of the six qualities, tranquil- does produce an auspicious influ- 
lity, &c. ; and, iv., the desire for ence ? 

liberation. 4 i.e., a word s incapacity to con- 

2 It may be suTcha-janaka, but it vey a meaning without some other 
is not itself sukha. word to complete the construction. 


" But ought not a prayer for an auspicious commence 
ment to be put at the beginning of a Sastra, in order to 
lay the hosts of obstacles that would hinder the com 
pletion of the work which the author desires to begin, 
and also to observe the immemorial practice of the good, 
since it has been said by the wise, Those astras become 
widely famous which have auspicious commencements, 
auspicious middles, and auspicious endings, and their 
students have long lives and are invincible in disputa 
tion ? l Now the word atha implies auspiciousness, 
since there is a Smriti which says, 

" The word Om and the word atha, these two in the 
ancient time, 

" Cleaving the throat of Brahman, came forth ; there 
fore they are both auspicious. 

" Therefore let the word atha stand here as signifying 
auspiciousness/ like the word vriddhi used by Panini 
in his opening siitra vriddhir dd aich! " 2 This view, 
however, is untenable; since the very word atha t when 
heard, has an auspicious influence, even though it be 
employed to convey some other special signification, just 
as the hearing the sound of lutes, flutes, &c. [is aus 
picious for one starting on a journey]. If you still object, 
" How can the particle atha have any other effect, if it is 
specially used here to produce the idea that the meaning 
of the sentence is that a new topic is commenced ? " we 
reply that it certainly can have such other additional 
effect, just as we see that jars of water brought for some 
other purpose are auspicious omens at the commence 
ment of a journey. 3 Nor does this contradict the smriti, 

1 This is found with some varia- as " the second strengthening of a 
tions in the Mahabhashya (p. 7, vowel." 

Kielhorn s ed.) 3 In the old Bengali poem Chan- 

2 The commentators hold that the di, we have an interesting list of 
word vriddliih is placed at the be- these omens. The hero Chandra- 
ginning of the first sutra, while ketu, starting on a journey, has the 
gunah in the second is placed at the following good omens : On his right 
end (ad en gunah}, in order to ensure hand a cow, a deer, a Brahman, a 
an auspicious opening, vriddhi mean- full-blown lotus ; on his left, a jackal 
ing "increase," "prosperity," as well and a jar full of water. He hears 


since the smriti will still hold good, as the words " they 
are both auspicious " mean only that they produce an 
auspicious effect. 

Nor can the particle atha have here the meaning of 
" reference to a previous topic," since the previously men 
tioned faults will all equally apply here, as this meaning 
really involves that of " after " [which we have already dis 
cussed and rejected]. And again, in such discussions as 
this, as to whether this particular atha means "the inceptive 
now " or " after," if another topic had been previously sug 
gested, then " reference thereto " would be a possible mean 
ing ; but in the present case [where no other topic has been 
previously suggested] it is not a possible meaning. There 
fore, by exhaustion, the commentator finally adopts, for 
the atka of the sutra, the remaining meaning of " the 
inceptive now." So, when it is said [in the Tandya Brah- 
mana, xvi. 8, I ; xvi. 10, i], " Now this is the Jyotis," 
" Now this is the Visvajyotis," x the particle atha is 
accepted as signifying the commencement of the descrip 
tion of a particular sacrifice, just as the atha in the 
commencement of the Mahabhashya, " now conies the 
exposition of words," signifies the commencement of the 
Institutes of Grammar. This has been declared by 
Vyasa in his Commentary on the Yoga Aphorisms, 
" the atha in this opening aphorism indicates a com 
mencement;" and Vachaspati has similarly explained it 
in his gloss ; therefore it may be considered as settled 
that the atha here indicates a commencement and also 
signifies auspiciousness. Therefore, accepting the view 

on his right hand the sound of fire omen according to all sdstras, and 
and a cowherdess calling " milk " to so is a tortoise, a rhinoceros, the 
buyers. He sees a cow with her calf, tuberous root of the water-lily, and 
a woman calling " ja,y&," diirvd grass, a hare." Elsewhere, a vulture, a 
rice, garlands of flowers, diamonds, kite, a lizard, and a woodman carry- 
sapphires, pearls, corals ; and on the ing wood are called bad omens, 
left twel\ 7 e women. He hears drums x These are the names of two out 
and cymbals, and men dancing and of the four sacrifices lasting for one 
singing " Hari." It is, however, all day, in which a thousand cows are 
spoiled by seeing a guana (godhikd). given to the officiating Brdhinans. 
The author adds, "This is a bad 


that this atha implies a commencement, let the student be 
left in peace to strive after a successful understanding of 
the sastra through the attainment of the yoga, which is 
its proposed subject, by means of the teacher s explana 
tion of its entire purport. But here some one may say, 
"Does not the smriti of Yajnavalkya say, Hiranyagarbha 
is the promulgator of the Yoga, and no other ancient 
sage ? how then is Patanjali the teacher thereof ? " We 
reply that it was for this reason that the venerable Patan 
jali, 1 that ocean of compassion, considering how difficult 
it was to" grasp all the different forms of Yoga scattered up 
and down in the Puranas, &c., and wishing to collect 
together their essence, commenced his anusdsana, the 
preposition anu implying that it was a teaching which 
followed a primary revelation and was not itself the 
immediate origin of the system. 

Since this atha in the aphorism signifies " commence 
ment," the full meaning of the sentence comes out as 
follows : " be it known that the institute for the exposi 
tion of the yoga is now commenced." In this institute 
the " object-matter," as being that which is produced by 
it, is yoga [or the " concentration of the mind "], with its 
means and its fruit; the producing this is its inferior "end;" 
supreme absorption (kaivalya) is the highest " end " of the 
yoga when it is produced. The " connection " between 
the institute and yoga is that of the producer and the 
thing to be produced ; the " connection " between yoga 
and supreme absorption is that of the means and the 
end ; and this is well known from Sruti and Smriti, 
as I have before shown. And it is established by the 
general context that those who aim at liberation are the 
duly qualified persons to hear this institute. Nor need 
any one be alarmed lest a similar course should be 
adopted with the opening aphorism of the Yedanta sutras, 
" Now, therefore, there is a wish to know Brahman ; " and 

1 He is here called phanipati, thor of the Malmbhashya, being re- 
" lord of snakes," Patanjali, the au- presented as a snake in mythology. 


lest here, too, we should seek to establish by the general 
context that all persons who aim at liberation are duly 
qualified students of the Vedanta. For the word atha, as 
there used, signifies " succession" [or," after"] ; and it is a 
settled point that the doctrine can only be transmitted 
through a regular channel to duly qualified students, and 
consequently the question cannot arise as to whether any 
other meaning is suggested by the context. Hence it has 
been said, " When Sruti comes [as the determining autho 
rity] the subject-matter and the rest have no place." l 
The full meaning of this is as follows : Where a thing is 
not apprehended from the Veda itself, there the " subject- 
matter" and the rest can establish the true meaning, not 
otherwise ; but wherever we can attain the meaning by a 
direct text, there the other modes of interpretation are 
irrelevant. For when a thing is declared by a text of the 
Veda which makes its meaning obvious at once, the " sub 
ject-matter" and the rest either establish a contrary con 
clusion or one not contrary. Now, in the former case, the 
authority which would establish this contrary conclusion 
is [by the very nature of " sruti "] already precluded from 
having any force ; and in the latter it is useless. This is 
all declared in Jaimini s aphorism [iii. 3, 14] ; "A definite 
text, a sign, the sentence, the subject-matter, the 
< relative position, or the title, when any of these come 
into collision, the later in order is the weaker because its 
meaning is more remote " 2 [and therefore less obvious]. 
It has been thus summed up 

1 Cf Sankara, Veddnta^Sut., iii. must be a liquid like ghee, since a 

ladle could not divide solid things 

~ This is the Mimdmsd rule for like the baked flour cakes. 3. 

settling the relative value of the Vdkya, "the being mentioned in 

proofsthat one thing is ancillary to one sentence," i.e., the context, 

another. I. Sruti, " a definite text," as in the text " (I cut) thee for 

as "let him offer with curds, "where food, thus saying, he cuts the 

curds are clearly an ancillary part of branch ;" here the words " (I cut) 

the sacrifice. 2. Linga, " a sign," or thee for food " are ancillary to the 

" the sense of the words, " as leading action of cutting ; or m the text, " I 

to an inference, as in the text " he offer the welcome (oblation) to 

divides by the ladle ;" here we in- Agni," the words "the welcome 

fer that the thing to be divided (oblation) to Agni, as they form 


< A text always precludes the rest ; the title is always 
precluded by any of the preceding modes ; 

"But whether any intervening one is precluded, or 
itself precludes, depends on circumstances." 

Therefore [after all this long discussion] it may be now 
considered as settled that, since it has an " object," as well 
as the other preliminaries, the study of the Sastra, which 
teaches the Yoga, is to be commenced like that of the 
Vedanta, which discusses the nature of Brahman. " But," 
it may be objected, " it is the Yoga which was said to be 
the object-matter, since it is this which is to be produced, 
not the Sastra." We grant that the Yoga is the principal 
object, as that which is to be produced ; but since it is 
produced by the Sastra, especially directed thereto, this 
Sastra is the means for its production, and, as a general 
rule, the agent s activity is directly concerned with the 
means rather than with the end. Just as the operations 
of Devadatta the woodcutter, i.e., his lifting his arm up 
and down, &c., relate rather to the instrument, i.e., the 
axe, than to the object, i.e., the tree, so here the speaker, 
Patanjali, in his immediate action of speaking, means 
the Yoga-Sastra as his primary object, while he intends 
the Yoga itself in his ultimate action of "denotation." 
In consequence of this distinction, the real meaning is 
that the commencing the Yogasastra is that which primarily 

one sentence with the words " I divine work," in connection with the 

offer," are ancillary to the act of mention of the sdnndyya vessels, 

offering. 4. Prakarana, " the sub- where this position proves that the 

ject-matter viewed as a whole, with hymn is ancillary to the action of 

an interdependence of its parts," as sprinkling those vessels. 6. Samd- 

in the darsa-purnamdsa sacrifice, khyd, " title ; " thus the Yajurveda 

where the praydja ceremonies, which is called the special book for the 

have no special fruit mentioned, adhvaryu priests ; hence in any rite 

produce, as parts, a mystic influ- mentioned in it they are primd 

ence (apurva) which helps forward facie to be considered "as the priests 

that influence of the whole by which employed. The order in the aphor- 

the worshippers obtain heaven, ism represents the relative weight 

Here the pra&arana proves them to to be attached to each ; the first, 

be ancillary. 5. Sthdna (or krama), sruti, being the most important ; the 

"relative position" or "order," as last, samdkhyd, the least. Of. Jai- 

the recital of the hymn Sundha- mini s Sutras, iii. 3, 14 ; Mimdmsd- 

dhvam, &c., " Be ye purified for the paribhdshd, pp. 8, 9. 


claims our attention ; while the " yoga," or the restraint of 
the modifications of the mind, is what is to be expounded 
in this Sastra, " But as we read in the lists of roots that 
the root yuj is used in the sense of joining/ should not the 
word yoga, its derivative, mean conjunction, and not re 
straint ? And indeed this has been said by Yajnavalkya : l 
The conjunction of the individual and the supreme 

souls is called yoga! " 

This, however, is untenable, since there is no possibility 
of any such action, 2 &c., in either as would produce this 
conjunction of the two souls. [Nor, again, is such an 
explanation needed in order to remove the opposition of 
other philosophical schools] ; for the notion of the con 
junction of two eternal things is opposed to the doctrines 
of the Vaiseshika and Nyaya schools [and therefore they 
would still oppose our theory]. And even if we accepted 
the explanation in accordance with the Mimamsa [or 
Vedanta], our Yogasastra would be rendered nugatory by 
this concession [and the very ground cut from under our 
feet] ; because the identity of the individual and supreme 
souls being in that school something already accomplished, 
it could not be regarded as something to be produced by 
our Sastra. And lastly, as it is notorious that roots are 
used in many different senses, the root yuj may very well 
be used here in the sense of " contemplation." 3 Thus it 
has been said 

" Particles, prepositions, and roots these three are all 

held to be of manifold meaning ; instances found in 

reading are their evidence." 

Therefore some authors expressly give yuj in this sense, 
and insert in their lists " yuj in the sense of samddhi" 
Nor does this contradict Yajnavalkya s declaration, as 
the word yoga, used by him, may bear this meaning ; and 
he has himself said 

1 I.e., Yogi-Ydjiiavalkya, the au- Icriyd, which properly belongs only 
thor of the Ydjnavalkya-gitd. See to the body, as the soul is drashtri. 
Hall, BiU. Index, p. 14 ; Aufrecht, 3 Sett, samddhi, or the restraining 
Bodl. Catal., p. 87 b. the mind and senses to profound 

2 Karman seems here used for contemplation. 


" Samddhi is the state of identity of the individual and 
supreme souls ; this abiding absolutely in Brahman 
is the samddhi of the individual soul." 
It has been also said by the venerable Vyasa [in his Com 
mentary on the Yoga-siitras, i. i], " Yoga is samddhi." 

An objection, however, may be here raised that "the 
term samddhi is used by Patanjali [in ii. 29] in the sense 
of one of the eight ancillary parts l of the eightfold con 
centration (or yoga) ; and the whole cannot be thus itself 
a part as well as a whole, since the principal and the 
ancillary must be completely different from each other, as 
all their attendant circumstances must be different, just as 
we see in the darsapurnamdsa sacrifices and their ancillary 
rites the pmyajas, and therefore samddhi cannot be the 
meaning of yoga! We however reply that this objection 
is incorrect ; for although the term samddhi is used for 
etymological reasons 2 to express the ancillary part which 
is really defined [in iii. 3] as " the contemplation which 
assumes the form of the object, and is apparently devoid of 
any nature of its own;" still the further use of this term to 
describe the principal state is justified by the author s 
wish to declare the ultimate oneness of the two states [as 
the inferior ultimately merges into the superior]. Nor 
can you hold that etymology alone can decide where a 
word can be used ; because if so, as the word go, " a bull/ 
is derived by all grammarians from the root gam, " to go," 
we ought never to use the phrase " a standing bull " [as 
the two words would be contradictory], and the man 
Devadatta, when going, would properly be called go, " a 
bull ; " and, moreover, the Sutra, i. 2, distinctly gives us 
a definite justification for employing the word in this 
sense when it declares that " concentration (yoga) is the 
suppression of the modifications of the thinking principle." 
[The second or principal sense of samddhi will therefore 
be quite distinct from the first or inferior.] 

1 Scil. "forbearance, religious ob- plation, and meditation (samddhi)." 
servance, postures, suppression of the 2 See Bhoja, Comm. iii. 3, samyay 
breath, restraint, attention, contem- ddldyate mano yatra sa samddhi h. 


" But surely if yoga is held to be the suppression of the 
modifications of the thinking principle, then as these modi 
fications abide in the soul as themselves partaking of the 
nature of knowledge, their suppression, or in other words 
their destruction/ would also abide in the soul, since it is a 
principle in logic that the antecedent non-existence and de 
struction abide in the same subject as the counter-entity to 
these negations ; ] and consequently in accordance with the 
maxim, This newly produced character will affect the sub 
ject in which it resides, the absolute independence of the 
soul itself would be destroyed." This, however, we do not 
allow ; because we maintain that these various modifica 
tions which are to be hindered, 2 such as " right notion," 
"misconception," "fancy," "sleep," and "memory" (i. 6), 
are attributes of the internal organ (cJiitta), since the power 
of pure intelligence, which is unchangeable, cannot become 
the site of this discriminative perception. Nor can you 
object that this unchangeable nature of the intelligent 
soul 3 has not been proved, since there is an argument to 
establish it; for the intelligent soul must be unchange 
able from the fact that it always knows, while that 
which is not always knowing is not unchangeable, as the 
internal organ, &c. And so again, if this soul were sus 
ceptible of change, then, as this change would be occa 
sional, we could not predicate its always knowing these 
moditications. But the true view is, that while the 
intelligent soul always remains as the presiding witness, 
there is another essentially pure substance 4 which abides 
always the same ; and as it is this which is affected by 
any given object, so it is this perceptible substance which 
is reflected as a shadow on the soul, and so produces an 

1 Thus, e.g., the antecedent non- 2 I read niroddhavydndm for niro- 

existence and the destruction of the dhdndm. 

pot are found in the two halves in 3 Chit - sakti and chiti sakti 

which the pot itself (the counter- soul. 

entity to its own non-existence) re- 4 The sattva of the buddhi or the 

sides by intimate relation (samavdya- internal organ. 


impression ; l and thus Soul itself is preserved in its own 
proper independence, and it is maintained to be the 
always knowing, and no suspicion of change alights upon 
it. That object by which the understanding becomes 
affected is known ; that object by which it is not affected 
is not known ; for the understanding is called "susceptible 
of change," because it resembles the iron, as it is suscep 
tible of being affected or not by the influence or want of 
influence of the object which resembles the magnet, this 
influence or want of influence producing respectively 
knowledge or the want of knowledge. " But inasmuch as 
the understanding and the senses which spring from egoism 
are all-pervading, are they not always connected with 
all objects, and thus would it not follow that there should 
be a knowledge everywhere and always of all things ? " 
We reply that even although we grant that they are all- 
pervading, it is only where a given understanding has 
certain modifications in a given body, and certain objects 
are in a connection with that body, that the knowledge of 
these objects only, and none other, is produced to that 
understanding ; and therefore, as this limitation is abso 
lute, we hold that objects are just like magnets, and 
affect the understanding just as these do iron, coming- 
in contact with it through the channels of the senses. 
Therefore, the " modifications " belong to the understanding, 
not to the soul ; and so says the Sruti, " Desire, volition, 
doubt, faith, want of faith, firmness, want of firmness, 
all this is only the mind." Moreover, the sage Panchasikha 
declared the unchangeable nature of the intelligent soul, 
" The power that enjoys is unchangeable ; " and so Pat- 
aiijali also (iv. 18), "The modifications of the under 
standing are always known, this arises from the un- 
changeableness of the Euling Soul." The following is 
the argument drawn out formally to establish the change- 

1 This second substance, " mind " the image of the object on a second 
or " understanding " (buddhi, chitta), looking-glass (nc. soulj. 
is like a looking-glass, which reflects 


ableness of the understanding. The understanding is 
susceptible of change because its various objects are now 
known and now not known, just like the organ of hear 
ing and the other organs of sense. Now, this change is no 
toriously threefold, i.e., a change of "property," of "aspect," 1 
and of " condition." When the subject, the understanding, 
perceives the colour "blue," &c., there is a change of 
" property" just as when the substance "gold" becomes a 
bracelet, a diadem, or an armlet ; there is a change of " as 
pect" when the property becomes present, past, or future ; 
and there is a change of " condition " when there is a mani 
festation or non-manifestation 2 of the perception, as of blue, 
&c. ; or, in the case of gold, the [relative] newness or oldness 
[at two different moments] would be its change of condi 
tion. These three kinds of change must be traced out by 
the reader for himself in different other cases. And thus 
we conclude that there is nothing inconsistent in our 
thesis that, since " right notion " and the other modifica 
tions are attributes of the understanding, their " suppres 
sion " will also have its site in the same organ. 

[Our opponent now urges a fresh and long objection 
to what we have said above.] " But if we accept your 
definition that * yoga is the suppression of the modifica 
tions of the chitta! this will apply also to sound sleep, 
since there too we may find the suppression [or suspen 
sion] of the modifications found in kshipta, vikshipta, 
mudlmf &c. ; but this would be wrong, because it is im 
possible for the afflictions to be abolished so long as 
those states called kshipta, &c., remain at all, and because 
they only hinder the attainment of the summum lonum. 
Let us examine this more closely. For the understand 
ing is called kshipta, restless/ when it is restless [with 

1 Vachaspati explains lalshana as of the lalsliana-parindma. Cf. the 
Tcdlablieda. Commentaries on iii. 1 3. 

2 I take ddi as meaning asphu- 3 These are generally called _ the 
tatva. The change of state takes five states of the thinking principle, 
place between the several moments chittabhumayas OTC avasthds. Cf. Com 
mentary, i. 2, 1 8. 


an excess of the quality rajas], as being tossed about 
amidst various objects which engage it. It is called mudha, 
blinded/ when it is possessed by the modification sleep 
and is sunk in a sea of darkness [owing to an excess of the 
quality tamas]. It is called vikshipta, unrestless, when 
it is different from the first state 1 [as filled with the 
quality sattva\. We must here, however, note a distinction ; 
for, in accordance with the line of the Bhagavad Gita (vi. 
34), The mind, Krishna, is fickle, turbulent, violent, 
and obstinate, the mind, though naturally restless, may 
occasionally become fixed by the transient fixedness of its 
objects ; but restlessness is innate to it, or it is produced 
in it by sickness, &c., or other consequences of former 
actions ; as it is said [in the Yoga Sutras, i. 30], Sickness, 
languor, doubt, carelessness, laziness, addiction to objects, 
erroneous perception, failure to attain some stage, and 
instability, these distractions of the mind are called 
obstacles . Here sickness means fever, &c., caused 
by the want of equilibrium between the three humours ; 
languor is the mind s want of activity ; doubt is a 
sort of notion which embraces two opposite alternatives ; 
carelessness is a negligence of using the means for 
producing meditation ; laziness is a want of exertion 
from heaviness of body, speech, or mind ; addiction to 
objects is an attachment to objects of sense ; erroneous 
perception is a mistaken notion of one thing for another; 
failure to attain some stage is the failing for some 
reason or other to arrive at the state of abstract medita 
tion ; instability is the mind s failure to continue there, 
even when the state of abstract meditation has been 
reached. Therefore we maintain that the suppression of 
the mind s modifications cannot be laid down as the defi 
nition of yoga! 

We reply, that even although we allow that, so far as 
regards the three conditions of the mind called kshipta, 

1 These three conditions respectively characterise men, demons, and gods. 


mudha, and wkshipta, which [as being connected with, 
the three qualities] are all to be avoided as faulty states, 
the suppression of the modifications in these conditions is 
itself something to be avoided [and so cannot be called 
yoga], this does not apply to the other two conditions 
called ekdgra and niruddha, which are to be pursued and 
attained ; and therefore the suppression of the modifica 
tions in these two praiseworthy conditions is rightly to 
be considered as yoga. Now by ekdgra we mean that 
state when the mind, entirely filled with the sattva, 
quality, is devoted to the one object of meditation ; and 
by niruddha we mean that state when all its develop 
ments are stopped, and only their latent impressions [or 
potentialities] remain. 

Now this samddhi, " meditation " [in the highest sense], 
is twofold: "that in which there is distinct recognition" 
(samprajndta), and " that in which distinct recognition 
is lost" (asamprajndtd) [Yoga S., i. 17, iS]. 1 The former 
is defined as that meditation where the thought is intent 
on its own object, and all the " modifications," such 
as " right notion," &c., so far as they depend on external 
things, are suppressed, or, according to the etymology of the 
term, it is where the intellect 2 is thoroughly recognised 
(samyak prajndyate) as distinct from Nature. It has a four 
fold division, as samtarJca, savichdra, sdnanda, and sdsmita. 
Now this " meditation " is a kind of "pondering" (bhdvand), 
which is the taking into the mind again and again, to the 
exclusion of all other objects, that which is to be pon 
dered. And that which is thus to be pondered is of two 
kinds, being either fswara or the twenty-five principles. 
And these principles also are of two kinds senseless and 
not senseless. Twenty-four, including nature, intellect, 
egoism, &c., are senseless; that which is not senseless is Soul. 
Now among these objects which are to be pondered, when, 
having taken as the object the gross elements, as earth, 

1 Much of this is taken from borrowed Ballantyne s translation. 
Bhoja s Commentary, and I have 2 Can cldtta mean " soul " here ? 


&c., pondering is pursued in the form of an investigation 
as to which is antecedent and which consequent, 1 or in 
the form of a union of the word, its meaning, and the 
idea which is to be produced [cf. i. 42] ; then the medita 
tion is called "argumentative (savitarka). When, having 
taken as its object something subtile, as the five subtile 
elements and the internal organ, pondering is pursued in 
relation to space, time, &c., then the meditation is called 
"deliberative" (savichdra). When the mind, commingled 
with some "passion" and " darkness," is pondered, then the 
meditation is called "beatific" (sdnanda), because "good 
ness " is then predominant, which consists in the mani 
festation of joy. 2 When pondering is pursued, having as 
its object the pure element of "goodness," unaffected by 
even a little of " passion " or " darkness," then that medita 
tion is called " egoistical " (sdsmita), because here personal 
existence 3 only remains, since the intellectual faculty 
becomes now predominant, and the quality of "goodness" 
has become quite subordinate [as a mere stepping-stone to 
higher things]. 

But the " meditation, where distinct recognition is lost," 
consists in the suppression of all " modifications " whatever. 

" But " [it may be asked] " was not concentration 
defined as the suppression of all the modifications ? How, 
then, can the meditation where there is distinct recogni 
tion be included in it at all, since we still find active in 
it that modification of the mind, with the quality of goodness 
predominant, which views the soul and the quality of good 
ness as distinct from each other?" This, however, is un 
tenable, because we maintain that concentration is the sup 
pression of the " modifications " of the thinking power, as 
especially stopping the operation of the " afflictions," the 
"actions," the "fructifications," and the "stock of deserts." 4 

1 I.e., as, e.g., whether the senses 3 In p. 164, line 2 infra, read 
produce the elements or the elements sattdmdtra for sattva-. Bhoja well 
the senses, &c. distinguishes asmitd from ahamkara. 

2 In p. 164, line 4 infra, read 4 For these see infra, and cf. Yoga 
sukhaprakdsamayasya. S., ii. 3, 12, 13. 


The "afflictions" (klesa) are well known as five, viz., 
ignorance, egoism, desire, aversion, and tenacity of mun 
dane existence. " But here a question is at once raised, In 
what sense is the word avidyd, "ignorance," used here ? Is 
it to be considered as an avyayibhdva compound, where the 
former portion is predominant, as in the word " above- 
board"? 1 or is it a tatpurusha [or karmadhdrayd] com 
pound, where the latter portion is predominant, as in the 
word " town- clerk " ? or is it a bahuvrihi compound, where 
both portions are dependent on something external to the 
compound, as " blue-eyed " ? It cannot be the first ; for if 
the former portion of the compound were predominant, then 
we should have the negation the emphatic part in avidyd 
(i.e., it would be an instance of what is called the express 
negation, or prasajya-pratishedha) ; 2 and consequently, as 
avidyd would be thus emphatically a negation, it would be 
unable to produce positive results, as the " afflictions," &c., 
and the very form of the word should not be feminine, but 
neuter. It cannot be the second ; for any knowledge, what 
ever thing s absence it may be characterised by (a -f- wdyd\ 
opposes the " afflictions," &c., and cannot therefore be their 
source. Nor can it be the third ; for then, in accordance 
with the words of the author of the Vritti, 3 " there is a 
~balmvrihi compound which is formed with some word 
meaning existence used after not, with the optional 
elision of this subsequent word" 4 we must explain this 
supposed baliuvrihi compound avidyd as follows : " That 
luddhi is to be characterised as avidyd (sc. an adjective), 

1 I have ventured to alter the (a.) "Not a drum was heard, not a 
examples, to suit the English trans- funeral note." 

lation. (b. ) " Tin watched the garden bough 

2 Where the negation is promi- shall sway." 

nent it is called prasajya-prati- The former corresponds to the logi- 

shedha; but where it is not promi- cian s atyantdbhdva, the latter to 

nent, we have the paryuddsa nega- anyonydbhdva or bheda. 
tion. In the former the negative 3 Cf. the vdrttika in Siddhanta 

is connected with the verb ; in the Kaum., i. 401. 

latter it is generally compounded 4 Thus adhana stands for avidya- 

\vith some other word, as, e.g. mdnadhana, with vidyamdna omitted 

in the compound. 


of which there is not a vidyd existing." But this explana 
tion is untenable ; for such an avidyd could not become the 
source of the "afflictions;" 1 and yet, on the other hand, 
it ought to be their source, 2 even though it were associated 
with the suppression of all the " modifications," 3 and were 
also accompanied by that discriminative knowledge of the 
soul and the quality of goodness [which is found in the 
sdsmita meditation]. 

" Now it is said [in the Yoga Sutras, ii. 4], " Ignorance is 
the field [or place of origin, i.e., source] of the others, whether 
they be dormant, extenuated, intercepted, or simple." They 
are said to be " dormant " when they are not manifested 
for want of something to wake them up ; they are called 
" extenuated " when, through one s meditatin" 1 on something 

o o o 

that is opposed to them, they are rendered inert ; they are 
called " intercepted " when they are overpowered by some 
other strong " affliction ; " they are called " simple " when 
they produce their several effects in the direct vicinity of 
what co-operates with them. This has been expressed by 
Vachaspati Misra, in his Gloss on Vyasa s Commentary, 
in the following memorial stanza : 

" The dormant afflictions are found in those souls which 
are absorbed in the tattvas [i.e., not embodied, but 
existing in an interval of mundane destruction] ; 
the extenuated 4 are found in yogins ; but the 
intercepted and the simple in those who are in 
contact with worldly objects." 

" No one proposes the fourth solution of the compound 
avidyd as a dvandva compound, 5 where both portions are 
equally predominant, because we cannot recognise here 
two equally independent subjects. Therefore under any 

1 As its subject would confessedly 4 I read tanvavastJidkJia with the 
be buddhi. printed edition of Vachaspati s Gloss. 

2 As it is avidyd after all. If tanudagdhdscha is correct, it must 

3 In p. 165, lines 16, 17, read (with mean tanutvena dagdkdh. 

my MS. of Vachaspati s Gloss), 5 As in rdmalakshmanau, Rtiina 

sarvavrittinirodhasampanndyd api and Lakshmana. 


one of these three admissible alternatives 1 the common 
notion of ignorance as being the cause of the afflictions 
would be overthrown." 

[We do not, however, concede this objector s view], 
because we may have recourse to the other kind of nega 
tion called paryuddsa [where the affirmative part is em 
phatic], and maintain that avidyd means a contradictory 
[or wrong} kind of knowledge, the reverse of vidyd ; and 
so it has been accepted by ancient writers. Thus it has 
been said 

" The particle implying negation does not signify ab 
sence [or non-existence ] when connected with 
a noun or a root ; thus the words abrdhmana and 
adharma respectively signify, what is other than 
a Brahman and what is contrary to justice. " 
And again 

" We are to learn all the uses of words from the custom 
of the ancient writers ; therefore a word must not 
be wrested from the use in which it has been 
already employed." 

Vachaspati also says, 2 "The connection of words and 
their meanings depends on general consent for its cer 
tainty ; and since we occasionally see that a tatpuruslia 
negation, where the latter portion is properly predominant, 
may overpower the direct meaning of this latter portion 
by its contradiction of it, we conclude that even here too 
[in avidyd} the real meaning is something contrary to 
vidyd " [i.e., the negative " non-knowledge " becomes ulti 
mately the positive " ignorance " 3 ]. It is with a view to 
this that it is said in the Yoga Aphorisms [ii. 5], " Ignor 
ance is the notion that the non-eternal, the impure, pain, 
and the non-soul are (severally) eternal, pure, pleasure, 
and soul." Viparyaya, "misconception," is denned as 

1 I read pakshatraye for paksha- nor, on the other hand, a "non- 
dvaye. friend," but something positive, an 

2 In his Comm. on Sut., ii. 5. " enemy." So agoshpada is said to 

3 Thus inimicus is not a "friend," mean "a forest." 


" the imagining of a thing in what is not that thing/ l [i.e., 
in its opposite] ; as, for instance, the imagining the " eter 
nal " in a " non-eternal " thing, i.e., a jar, or the imagin 
ing the " pure " in the " impure " body, 2 when it has been 
declared by a proverbial couplet 3 

"The wise recognise the body as impure, from its 
original place [the womb], from its primal seed, 
from its composition [of humours, &c.], from per 
spiration, from death [as even a Brahman s body 
denies], and from the fact that it has to be made 
pure by rites." 

So, in accordance with the principle enounced in the 
aphorism (ii. 15), "To the discriminating everything is 
simply pain, through the pain which arises in the ultimate 
issue of everything, 4 or through the anxiety to secure 
it [while it is enjoyed], or through the latent impres 
sions which it leaves behind, and also from the mutual 
opposition of the influences of the three qualities " [in the 
form of pleasure, pain, and stupid indifference], ignor 
ance transfers the idea of "pleasure" to what is really 
" pain," as, e.g., garlands, sandal- wood, women, &c. ; and 
similarly it conceives the " non-soul/ e.g., the body, &c., 
as the " soul." As it has been said 

" But ignorance is when living beings transfer the 

notion of soul to the non-soul, as the body, &c. ; 

" This causes bondage ; but in the abolition thereof is 


Thus this ignorance consists of four kinds. 5 
" But [it may be objected] in these four special kinds 
of ignorance should there not be given some general defi 
nition applying to them all, as otherwise their special 

1 Cf. Yoga Sui., i. 8. his explanation of it ; he calls it 

2 In p. 1 66, line 4 infra, read vaiydsalci gdthd. 

Tcdyddau for Tcdryddau. 4 Since the continued enjoyment 

3 This couplet is quoted by Vya"sa of an object only increases the desire 
in his Comm. on Yoga Sutras, ii. 5, for more, and its loss gives correspon- 
and I have followed Vachaspati in dent regret (cf. Bhag. G. xviii. 38). 

5 Literally, " it has four feet." 


characteristics cannot be established ? For thus it has 
been said by Bhatta Kumarila 

Without some general definition, a more special defi 
nition cannot be given by itself ; therefore it must 
not be even mentioned here. " 

This, however, must not be urged here, as it is sufficiently 
met by the general definition of misconception, already ad 
duced above, as " the imagining of a thing in its opposite." 

" Egoism " (asmitd) is the notion that the two separate 
things, the soul and the quality of purity, 1 are one and the 
same, as is said (ii. 6), " Egoism is the identifying of the 
seer with the power of sight." " Desire " (rdga) is a long 
ing, in the shape of a thirst, for the means of enjoyment, 
preceded by the remembrance of enjoyment, on the part of 
one who has known joy. "Aversion" (dvesha) is the feel 
ing of blame felt towards the means of pain, similarly pre 
ceded by the remembrance of pain, on the part of one who 
lias known it. This is expressed in the two aphorisms, 
" Desire is what dwells on pleasure ; " " Aversion is what 
dwells on pain " (ii. 7, 8). 

Here a grammatical question may be raised, " Are we 
to consider this word anusayin ( dwelling ) as formed 
by the krit affix nini in the sense of what is habitual, 
or the taddhita affix ini in the sense of matup ? It cannot 
be the former, since the affix nini cannot be used after 
a root compounded with a preposition as anusi ; for, as 
the word supi has already occurred in the Sutra, iii. 2, 4, 
and has been exerting its influence in the following sutras, 
this word must have been introduced a second time in the 
Sutra, iii. 2, 78, supy afdtau ninis tdclichJiilye? on purpose 
to exclude prepositions, as these have no case termina 
tions ; and even if we did strain a point to allow them, still 
it would follow by the Sutra, vii. 2, 115, aclw nniti, 3 that 

1 Thus " sight," or the power of a root in the sense of what is habitual, 

seeing, is a modification of the qua- when the upapada, or subordinate 

lity of sattva unobstructed by rajas word, is not a word meaning genus 

and tamos. and ends in a case." 

z " Let the affix nini be used after 3 " Let vriddhi be the substitute 


the radical vowel must be subject to vriddhi, and so the 
word must be anusdyin, in accordance with the analogy 
of such words as atisdyin, &c. Nor is the latter view 
tenable (i.e., that it is the taddhita affix mi 1 ), since ini is 
forbidden by the technical verse 

These two affixes 2 are not used after a monosyllable 
nor a krit formation, nor a word meaning genus/ 
nor with a word in the locative case ; 
and the word anusaya is clearly a krit formation as it ends 
with the affix ach 3 [which brings it under this prohibition, 
and so renders it insusceptible of the affix ini]. Conse 
quently, the word anusayin in the Yoga aphorism is one 
the formation of which it is very hard to justify." 4 This 
cavil, however, is not to be admitted ; since the rule is 
only to be understood as applying generally, not abso 
lutely, as it does not refer to something of essential im 
portance. Hence the author of the Vritti has said 
" The word iti, as implying the idea of popular accep 
tation, is everywhere connected with the examples 
of this rule 5 [i.e., it is not an absolute law]." 
Therefore, sometimes the prohibited cases are found, as 
kdryin, kdryika [where the affixes are added after a krit 
formation], tandulin, tandulika [where they are added 
after a word meaning "genus"]. Hence the prohibition is 
only general, not absolute, after krit formations and words 
meaning " genus," and therefore the use of the affix ini is 
justified, although the word anusaya is formed by a krit 
affix. This doubt therefore is settled. 

of a base ending in a vowel, when vdn ; (4.) dandavati said (i.e., dandd 

that which has an indicatory n or n asydm santi). 

follows ; " nini has an indicatory n. d By iii. 3, 56. 

1 Sc. anusaya + ini = anusayin. 4 It is curious to see the great 

2 Ini and than, which respectively grammarian s favourite study ob- 
leave in and ika ; thus danda gives truding itself here on such a slender 
dandin and dandika. The line is pretext. 

quoted by Boehtlingk, vol. ii. p. 2 1 7, 5 See the Kdsikd on Pdn. v. 2, 
on Pan. v. 2, 115, and is explained 115. For vivakskdrtha (meaning 
in the Kd&kd, ad loc. The different " general currency "), compare Corn- 
prohibitions are illustrated by the mentary on Pan. ii. 2, 27. The edi- 
examples: (i.) svavdn, khavdn ; (2.) tion in the Benares Pandit reads 
Icdrakavdn ; (3.) vydgliravdn, simha- mshayaniyamdriha. 


The fifth " affliction," called " tenacity of mundane 
existence " (abhinivesa), is what prevails in the case of 
all living beings, from the worm up to the philosopher, 
springing up daily, without any immediate cause, in the 
form of a dread, " May I not be separated from the body, 
things sensible, &c.," through the force of the impression 
left by the experience of the pain of the deaths which 
were suffered in previous lives, this is proved by uni 
versal experience, since every individual has the wish, 
" May I not cease to be," " May I be." This is declared 
in the aphorism, " Tenacity of mundane existence, flowing 
on through its own nature, is notorious even in the case of 
the philosopher " [ii. 9], These five, " ignorance," &c., are 
well known as the " afflictions " (klesa), since they afflict 
the soul, as bringing upon it various mundane troubles. 

[We next describe the karmdsaya of ii. 12, the "stock 
of works " or " merits " in the mind.] " Works " (karman) 
consist of enjoined or forbidden actions, as the jyotish- 
toma sacrifice, brahmanicide, &c. " Stock " (dsaya) is the 
balance of the fruits of previous works, which lie stored 
up in the mind in the form of " mental deposits " of merit 
or demerit, until they ripen in the individual soul s own 
experience as "rank," "years," and "enjoyment" [ii. 13]. 

JSTow " concentration " \}jocjd\ consists [by i. 2] in " the 
suppression of the modifications of the thinking principle," 
which stops the operation of the "afflictions," &c. ; and 
this " suppression " is not considered to be merely the non- 
existence of the modifications [i.e., a mere negation], 
because, if it were a mere negation, it could not produce 
positive impressions on the mind ; but it is rather the site 
of this non-existence, 1 a particular state of the thinking 
principle, called by the four names [which will be fully 
described hereafter], madhumati, madlmpratikd, visokd, 
and samskdraseshatd. The word nirodha thus corresponds 
to its etymological explanation as " that in which the modi 
fications of the thinking principle, right notion, miscon- 

1 i.e., Thus nirodha is not vrittcr abhdrafi, but abhdvasydsryah. 


ception, &c., are suppressed (nirudhyante). This suppres 
sion of the modifications is produced by " exercise " and 
" dispassion " [i. 12]. " Exercise is the repeated effort that 
the internal organ shall remain in its proper state " [i. 1 3]. 
This " remaining in its proper state " is a particular kind 
of development, whereby the thinking principle remains in 
its natural state, unaffected by those modifications which 
at different times assume the form of revealing, ener 
gising, and controlling. 1 " Exercise " is an effort directed 
to this, an endeavour again and again to reduce the in 
ternal organ to such a condition. The locative case, sthitau, 
in the aphorism is intended to express the object or aim, as 
in the well-known phrase, " He kills the elephant for 
its skin." 2 " Dispassion is the consciousness of having 
overcome desire in him who thirsts after neither the 
objects that are seen nor those that are heard of in reve 
lation" [i. 15]. "Dispassion" is thus the reflection, 
" These objects are subject to me, not I to them," in one 
who feels no interest in the things of this world or the 
next, from perceiving the imperfections attached to them. 

Now, in order to reduce the " afflictions " which hinder 
meditation and to attain meditation, the yogin must first 
direct his attention to practical concentration, and " exer 
cise " and " dispassion " are of especial use in its attain 
ment. This has been said by Krishna in the Bhagavad 
Gita [vi. 3]- 

" Action is the means to the sage who wishes to rise to 
yoga ; 

" But to him who has risen to it, tranquillity is said to 
be the means." 

Patanjali has thus defined the practical yoga : " Practical 
concentration is mortification, recitation of texts, and 
resignation to the Lord" [ii. i]. Yajnavalkya has de 
scribed " mortification " 

1 I read in p. 1 68, last line, prakdsapravrittiniyamartipa, from Bhoja s 
comment on i. 12. 

2 See Ktisika", ii. 3, 36. 



" By the way prescribed in sacred rule, by the difficult 

chandrayana fast, &c., 
" Thus to dry up the body they call the highest of all 

mortifications." 1 

" Eecitation of texts " is the repetition of the syllable 
Om, the gdyatri, &c. Now these mantras are of two kinds, 
Vaidik and Tantrik. The Yaidik are also of two kinds, 
those chanted and those not chanted. Those chanted are 
the sdmans; those not chanted are either in metre, i.e., 
the richas, or in prose, i.e., the yajumshi, as has been said 
by Jaimini, 2 " Of these, that is a rich in which by the force 
of the sense there is a definite division into pddas [or 
portions of a verse] ; the name sdman is applied to chanted 
portions ; the word yajus is applied to the rest." Those 
mantras are called Tantrik which are set forth in sacred 
books that are directed to topics of voluntary devotion ; 3 
and these are again threefold, as female, male, and neuter ; 
as it has been said 

" The mantras are of three kinds, as female, male, and 

neuter : 

" The female are those which end in the wife of fire 
(i.e., the exclamation svdhd) ; the neuter those 
which end in namas ; 
" The rest are male, and considered the best. They are 

all-powerful in mesmerising another s will, &c." 
They are called " all-powerful " (siddha) because they 
counteract all defects in their performance, and produce 
their effect even when the ordinary consecrating cere 
monies, as bathing, &c., have been omitted. 

Now the peculiar " consecrating ceremonies " (samsJcdra) 
are ten, and they have been thus described in the Sdradd- 

" There are said to be ten preliminary ceremonies which 
give to mantras efficacy : 

1 This passage probably occurs in 2 Mimdmsd, Sutras, ii. I, 35~37- 
the Ydjnavalkyagitd of Yogi-yajna- 3 The tantras are not properly 

valkya. See Colebrooke s Essays concerned with what is nit^a cr 

(ed. 2), vol. i. p. 145, note. t naimittika; they are kdmya. 


" These mantras are thus made complete ; they are 

thoroughly consecrated. 
"The begetting/ the vivifying/ the smiting/ the 


" The sprinkling-/ the ( purifying/ the fattening/ 
" The satisfying/ the illumining/ the concealing/ 

these are the ten consecrations of mantras. 
"The begetting (janana) is the extracting of the 

mantra from its vowels and consonants. 
" The wise man should mutter the several letters of the 

mantra, each united to Om, 
"According to the number of the letters. This they 

call the vivifying (jivand). 
"Having written the letters of the mantra, let him 

smite each with sandal-water, 
" Uttering at each the mystic seed of air. 1 This is 

called the smiting (tddana). 
" Having written the letters of the mantra, let him strike 

them with oleander flowers, 
"Each enumerated with a letter. This is called the 

awakening (lodhana). 
" Let the adept, according to the ritual prescribed in his 

own special tantra, 

" Sprinkle the letters, according to their number, with 
leaves of the Ficus religiosa. This is the sprink 
ling (dbhisheka). 
" Having meditated on the mantra in his mind, let him 

consume by the jyotir-mantra 
"The threefold impurity of the mantra. This is the 

purification (vimali-karana). 
" The utterance of the jyotir-mantra, together with Om, 

and the mantras of Vyoman and Agni, 
" And the sprinkling of every letter with water from a 

bunch of kusa grass, 

" With the mystical seed of water 2 duly muttered, this 
is held to be the fattening (apydyand). 

1 The vlja of air is the syllable jam. 

2 The vija of water is the syllable bam. 


" The satiating libation over the mantra with mantra- 

hallowed water is the satisfying (tarpana). 
" The joining of the mantra with Om and the seeds 
of Maya 1 and Kama 2 is called its illumining 
(Mpana). * 

"The non-publication of the mantra which is being 

muttered this is its concealing (gopana). 
" These ten consecrating ceremonies are kept close in 

all tantras ; 
"And the adept who practises them according to the 

tradition obtains his desire ; 

" And ruddha, Ulita, vichhinna, supta, sapta, and the rest, 
" All these faults in the mantra rites are abolished by 

these excellent consecrations." 

But enough of this venturing to make public the tantra 
mysteries connected with mantras, which has suddenly led 
us astray like an unexpected Bacchanalian dance. 1 

The third form of practical yoga, "resignation to the 
Lord" (tvara-pranidhdna),ia the consigning all one s works, 
whether mentioned or not, without regard to fruit, to the 
Supreme Lord, the Supremely Venerable. As it has been 

" Whatever I do, good or bad, voluntary or involuntary, 

" That is all made over to thee ; I act as impelled by thee." 

This self-resignation is also sometimes denned as " the 

surrender of the fruits of one s actions," and is thus a 

peculiar kind of faith, since most men act only with a 

selfish regard to the fruit. Thus it is sung in the Bhagavad 

Gita [ii. 47] 

" Let thy sole concern be with action and never with 

the fruits ; 
" Be not attracted by the fruit of the action, nor be thou 

attached to inaction." 

The harmfulness of aiming at the fruit of an action 
has been declared by the venerable JSTilakantha-bharati 

1 ffrim. 2 SWm., 

3 Tdndava is the frantic dance of the god Siva and his votaries. 


" Even a penance accomplished by great effort, but 
vitiated by desire, 

" Produces only disgust in the Great Lord, like milk 
which has been licked by a dog." 

Now this prescribed practice of mortification, recitation, 
and resignation is itself called yoga, because it is a 
means for producing yoga, this being an instance of the 
function of words called " superimponent pure Indication," 
as in the well-known example, " Butter is longevity." " In 
dication " is the establishing of another meaning of a word 
from the incompatibility of its principal meaning with the 
rest of the sentence, and from the connection of this new 
meaning with the former; it is twofold, as founded on 
notoriety or on a motive. This has been declared in the 
Kdvya-prakdsa [ii. 9] 

"When, in consequence of the incompatibility of the 
principal meaning of a word, and yet in connection 
with it, another meaning is indicated through noto 
riety or a motive, this is Indication, the super- 
added function of the word." 

Now the word "this" [i.e., tat in the neuter, which the 
neuter yat in the extract would have naturally led us to 
expect instead of the feminine sd] would have signified 
some neuter word, like " implying," which is involved as a 
subordinate part of the verb "is indicated." But sd is 
used in the feminine [by attraction to agree with lakshand], 
"this is indication," i.e., the neuter "this" is put in the 
feminine through its dependence on the predicate. This 
has been explained by Kaiyata, " Of those pronouns which 
imply the identity of the subject and the predicate, the 
former takes the gender of the former, the latter of the 
latter." l Now " expert (kusala) in business " is an example 
of Indication from notoriety ; for the word kusala, which is 

1 Literally "they take severally in providum, acutum, plenum rationis 

order the gender of one of the two." etconsilii, quern vocamus hominem," 

Cf . " Thebae ipsae quod Boeotise caput Cic., Legg, i. 7. 
est," Livy, xlii. 44 ; " Animal hoc 


significant in its parts by being analysed etymologically as 
kusam + ldti, " one who gathers kusa grass for the sacrifice," 
is here employed to mean "expert " through the relation of 
a similarity in character, as both are persons of discern 
ment;- and this does not need a motive any. more than 
Denotation does, since each is the using a word in its recog 
nised conventional sense in accordance with the immemorial 
tradition of the elders. Hence it has been said 

" Some instances of indication are known by notoriety 
from their immediate significance, just as is the 
case in denotation [the primary power of a 

Therefore indication based on notoriety has no regard 
to any motive. Although a word, when it is employed, 
first establishes its principal meaning, and then by that 
meaning a second meaning is subsequently indicated, and 
so indication belongs properly to the principal meaning and 
not to the word ; still, since it is superadded to the word 
which originally established the primary meaning, it is 
called [improperly by metonymy] a function of the word. 
It was with a view to this that the author of the Kavya- 
prakasa used the expression, " This is Indication, the 
superadded function of the word." But the indication based 
on a motive is of six kinds : i. inclusive indication, 1 as 
" the lances enter " [where we really mean " men with the 
lances "] ; 2. indicative indication, as " the benches shout " 
[where the spectators are meant without the benches] ; 3. 
qualified 2 superimponent indication, as " the man of the 
Panjab is an ox " [here the object is not swallowed up in 
the simile] ; 4. qualified introsusceptive indication, as 
" that ox " [here the man is swallowed up in the simile] ; 
5. pure superimponent indication, as " gM is life ;" 6. pure 

1 I have borrowed these terms from his stupidity ; pure indication 
from Ballantyne s translation of the from any other relation, as cause and 
Sahitya-darpana. effect, &c., thus butter is the c. iuse of 

2 Qualified indication arises from longevity, 
likeness, as the man is like an ox 


introsusceptive indication, as " verily this is life." This 
has been all explained in the Kavya-prakasa [ii. 10-12]. 
But enough of this churning of the depths of rhetorical 

This yoga has been declared to have eight things ancillary 
to it (anga) ; these are the forbearances, religious observ 
ances, postures, suppression of the breath, restraint, atten 
tion, contemplation, and meditation [ii. 29]. Pataiijali 
says, " Forbearance consists in not wishing to kill, veracity, 
not stealing, continence, not coveting " [ii. 30]. " Eeligious 
observances are purifications, contentment, mortification, 
recitation of texts, and resignation to the Lord" [ii. 
32]; and these are described in the Vishnu Purana [vi. 7, 


"The sage who brings his mind into a fit state for 

attaining Brahman, practises, void of all desire, 
" Continence, abstinence from injury, truth, non-steal 
ing, and non-coveting ; 
" Self-controlled, he should practise recitation of texts, 

purification, contentment, and austerity, 
"And then he should make his mind intent on the 

Supreme Brahman. 
" These are respectively called the five forbearances 

and the five religious observances ; 
"They bestow excellent rewards when done through 
desire of reward, and eternal liberation to those 
void of desire." 

" A posture is what is steady and pleasant " [ii. 46] ; 
it is of ten kinds, as the padma, bhadra, vira, svastika, 
dandaka, sopasraya, paryaiika, kraunchanishadana, ushtra- 
nishadana, samasamsthdna. Yajnavalkya has described 
each of them in the passage which commences 

" Let him hold fast his two great toes with his two 

hands, but in reverse order, 
" Having placed the soles of his feet, chief of Brah- 

mans, on his thighs ; 
" This will be the padma posture, held in honour by all." 


The descriptions of the others must be sought in that 
work. When this steadiness of posture has been attained, 
" regulation of the breath " is practised, and this consists 
in " a cutting short of the motion of inspiration and ex 
piration " [ii. 49]. Inspiration is the drawing in of the 
external air; expiration is the expelling of the air within 
the body ; and " regulation of the breath " is the cessa 
tion of activity in both movements. " But [it may be 
objected] this cannot be accepted as a general definition 
of regulation of breath/ since it fails to apply to the 
special kinds, as rechalca, puraka, and Jcumbhaka." We 
reply that there is here no fault in the definition, since the 
" cutting short of the motion of inspiration and expira 
tion " is found in all these special kinds. Thus rechaka, 
which is the expulsion of the air within the body, is 
only that regulation of the breath, which has been men 
tioned before as " expiration ; " and puraka, which is 
the [regulated] retention of the external air within the 
body, is the " inspiration ; " and kumbkaJca is the internal 
suspension of breathing, when the vital air, called prdna, 
remains motionless like water in a jar (kumbha). Thus 
the " cutting short of the motion of inspiration and ex 
piration " applies to all, and consequently the objector s 
doubt is needless. 

Now this air, beginning from sunrise, remains two 
ghatikds and a half l in each artery 2 (nddi), like the re 
volving buckets on a waterwheel. 3 Thus in the course 
of a day and night there are produced 21,600 inspirations 

1 I.e., an hour, a ghatiJcd being tras repeated with the offerings to 
twenty-four minutes. the seasons, is discussed. " The 

2 The nddis or tubular vessels are seasons never stand still ; following 
generally reckoned to be 101, with each other in order one by one, as 
ten principal ones ; others make spring, summer, the rains, autumn, 
sixteen principal nddis. They seem the cold and the foggy seasons, each 
taken afterwards in pairs. consisting of two months, and so 

3 Madhava uses the same illus- constituting the year of twelve 
tration in his commentary on the months, they continue revolving 
passage in the Aitareya Brahmana again and again like a waterwheel 
(iii. 29\ where the relation of the (ghatiycvntravat) ; hence the seasons 
vital airs, the seasons, and the man- never pause in their course." 


and expirations. Hence it has been said by those who 
know the secret of transmitting the mantras, concerning 
the transmission of the ajapdmantra l 

"Six hundred to Ganesa, six thousand to the self- 
existent Brahman, 

" Six thousand to Vishnu, six thousand to Siva, 
" One thousand to the Guru (Brihaspati), one thousand 

to the Supreme Soul, 
" And one thousand to the soul : thus I make over the 

performed muttering." 

So at the time of the passing of the air through the 
arteries, the elements, earth, &c., must be understood, 
according to their different colours, by those who wish to 
obtain the highest good. This has been thus explained 
by the wise 

" Let each artery convey the air two ghatis and a half 

from sunrise. 
" There is a continual resemblance of the two arteries 2 

to the buckets on a revolving waterwheel. 
" Nine hundred inspirations and expirations of the air 

take place [in the hour], 
" And all combined produce the total of twenty-one 

thousand six hundred in a day and night. 
" The time that is spent in uttering thirty -six guna 

letters, 3 
" That time elapses while the air passes along in the 

interval between two arteries. 

" There are five elements in each of the two conduct 
ing arteries, 

1 This refers to a peculiar tenet of 3 I cannot explain this. We 
Hindu mysticism, that each invo- might read guruvarndndm. for guna- 
luntary inspiration and expiration varndndm, as the time spent in 
constitutes a mantra, as their sound uttering a guruvarna is a vipala, 
expresses the word so ham (i.e., sixty of which make a pala, and two 
Tiamsah), " I am he." This mantra and a half palas make a minute ; but 
is repeated 21, 600 times in every this seems inconsistent with the other 
twenty-four hours ; it is called the numerical details. The whole pas- 
ajapdmantra, i.e., the mantra uttered sage may be compared with the 
without voluntary muttering. opening of the fifth act of the Mala- 

2 I.e., that which conveys the in- tirruidhava. 
haled and the exhaled breath. 


" They bear it along day and night ; these are to be 

known by the self-restrained. 
" Fire bears above, water below ; air moves across ; 
" Earth in the half -hollow ; ether moves everywhere. 
" They bear along in order, air, fire, water, earth, ether ; 
" This is to be known in its due order in the two con 
ducting arteries. 

" The palas l of earth are fifty, of water forty, 
" Of fire thirty, of air twenty, of ether ten. 
" This is the amount of time taken for the bearing ; but 

the reason that the two arteries are so disturbed 
" Is that earth has five properties, 2 water four, 
" Fire has three, air two, and ether one. 
" There are ten palas for each property ; hence earth has 

fifty palas, 
" And each, from water downwards, loses successively. 

Now the five properties of earth 
" Are odour, savour, colour, tangibility, and audibleness ; 

and these decrease one by one. 
" The two elements, earth and water, produce their 

fruit by the influence of quiet/ 
" But fire, air, and ether by the influence of brightness/ 

c restlessness/ and immensity. 3 
" The characteristic signs of earth, water, fire, air, and 

ether are now declared ; 

" Of the first steadfastness of mind ; through the cold 
ness of the second arises desire; 
" From the third anger and grief ; from the fourth 

fickleness of mind ; 
" From the fifth the absence of any object, or mental 

impressions of latent merit. 
" Let the devotee place his thumbs in his ears, and a 

middle finger in each nostril, 

1 Sixty palas make a ghatikd 2 Cf. Colebrooke s Essays, vol. i. 

(50 + 40 4- 30 + 20 + 10 = 150, i.e., p. 256. 

the palas in two and a half ghatikds 3 Literally "the being ever more." 
or one hour). 


"And the little finger and the one next to it in the 
corners of his mouth, and the two remaining fingers 
in the corners of his eyes, 
" Then there will arise in due order the knowledge of 

the earth and the other elements within him, 
"The first four hy yellow, white, dark red, and dark 

blue spots, 1 the ether has no symbol." 
When the element air is thus comprehended and its 
restraint is accomplished, the evil influence of works 
which concealed discriminating knowledge is destroyed 
[ii. 52]; hence it has been said 

"There is no austerity superior to regulation of the 

breath." 2 
And again 

"As the dross of metals, when they are melted, is con 

" So the serpents of the senses are consumed by regu 
lation of the breath." 3 

Now in this way, having his mind purified by the " for 
bearances" and the other things subservient to concen 
tration, the devotee is to attain " self-mastery " (samyama) 4 
and " restraint " (pratydhdra). " Restraint " is the accom 
modation of the senses, as the eye, &c., to the nature of the 
mind, 5 which is intent on the soul s unaltered nature, while 
they abandon all concernment with their own several ob 
jects, which might excite desire or anger or stupid indiffer 
ence. This is expressed by the etymology of the word; the 
senses are drawn to it (d + hri), away from them (pratipa). 
" But is it not the mind which is then intent upon the 
soul and not the senses, since these are only adapted for 
external objects, and therefore have no power for this 
supposed action ? How, therefore, could they be accommo- 

i For these colours cf. Chkdndogya 4 This is defined in the Yoga Sut , 

Uv viii. 6 ; Maitri Up., vi. 30. "i- 4, as consisting of the united 

* This is an anonymous quotation operation towards one object of con- 
in Vyasa s Comm. , templation, attention, and medita- 

s This seems a variation of Sloka tion. 

7 of the Amrita-ndda Up. See 5 I.e., the internal organ (chitta). 
Weber, Indische Stud., ix. 26. 


dated to the nature of the mind ? " What you say is quite 
true ; and therefore the author of the aphorisms, having 
an eye to their want of power for this, introduced the 
words "as it were," to express "resemblance." "Restraint 
is, as it were, the accommodation of the senses to the 
nature of the mind in the absence of concernment with 
each one s own object" [ii. 54]. Their absence of con 
cernment with their several objects for the sake of being 
accommodated to the nature of the mind is this " resem 
blance" which we mean. Since, when the mind is re 
strained, the eye, &c., are restrained, no fresh effort is to 
be expected from them, and they follow the mind as bees 
follow their king. This has been declared in the Yishnu- 
purana [vi. 7, 43, 44] 

" Let the devotee, restraining his organs of sense, which 
ever tend to pursue external objects, 

" Himself intent on restraint, make them conformable 
to the mind ; 

" By this is effected the entire subjugation of the un 
steady senses ; 

" If they are not controlled, the yogin will not accom 
plish his yoga! l 

"Attention" (dhdranti) is the fixing the mind, by with 
drawing it from all other objects, on some place, whether 
connected with the internal self, as the circle of the 
navel, the lotus of the heart, the top of the suskumnd 
artery, &c., or something external, as Prajapati, Vasava, 
Hiranyagarbha, &c. This is declared by the aphorism, 
" Attention is the fixing the mind on a place " [iii. i] ; 
and so, too, say the followers of the Puranas 

" By regulation of breath having controlled the air, and 
by restraint the senses, 

" Let him next make the perfect asylum the dwelling- 
place of his mind." 2 

i This couplet is corrupt in the 2 Vishnu-pur., vi. 7, 45, with one 

text. I follow the reading of the or two variations. The " perfect 

Bombay edition of the Purdna (only asylum " is Brahman, formless or 

reading in line 3 chcddtmandm). possessing form. 


The continual flow of thought in this place, resting on 
the object to be contemplated, and avoiding all incon 
gruous thoughts, is " contemplation " (dhydna) ; thus it 
is said, " A course of uniform thought there, is contem 
plation " [iii. 2\. Others also have said 

" A continued succession of thoughts, intent on objects 
of that kind and desiring no other, 

" This is l contemplation, it is thus effected by the 
first six of the ancillary things." 

We incidentally, in elucidating something else, dis 
cussed the remaining eighth ancillary thing, " meditation " 
(samddhi, see p. 243). By this practice of the ancillary 
means of yoga, pursued for a long time with uninterrupted 
earnestness, the " afflictions " which hinder meditation are 
abolished, and through " exercise " and " dispassion " the 
devotee attains to the perfections designated by the names 
Madhumati and the rest. 

" But why do you needlessly frighten us with unknown 
and monstrous words from the dialects of Karnata, 
Gauda, 1 and Lata ? " 2 We do not want to frighten you, 
but rather to gratify you by explaining the meaning of 
these strange words; therefore let the reader who is so 
needlessly alarmed listen to us with attention. 

i. The Madhumati perfection, this is the perfection of 
meditation, called " the knowledge which holds to the 
truth," consisting in the illumination of unsullied purity 
by means of the contemplation of " goodness," composed of 
the manifestation of joy, with every trace of " passion " or 
" darkness " abolished by " exercise," " dispassion," &c. 
Thus it is said in the aphorisms, " In that case there is 
the knowledge which holds to the truth " [i. 48]. It holds 
" to the truth," i.e., to the real ; it is never overshadowed 
by error. " In that case," i.e., when firmly established, there 
arises this knowledge to the second yogin. For the yogins 

1 The old name for the central and part of Guzerat ; it is the AapiKr/ 
part of Bengal. of Ptolemy. 

2 A country comprising Khandesh 


or devotees to the practice of yoga are well known to be 
of four kinds, viz., 

i. The prdtkamaJcalpika, in whom the light has just 
entered, 1 but, as it has been said, " he has not won the light 
which consists in the power of knowing another s thoughts, 
&c.;" 2. The madhubhtimiTca, who possesses the knowledge 
which holds to the truth ; 3. The prajndjyotis, who has 
subdued the elements and the senses ; 4. The atikrdnta- 
bhdvanfya, who has attained the highest dispassion. 

ii. The Madhupratika perfections are swiftness like 
thought, &c. These are declared to be " swiftness like 
thought, the being without organs, and the conquest of 
nature" [iii. 49]. "Swiftness like thought" is the attain 
ment by the body of exceeding swiftness of motion, like 
thought ; "the being without bodily organs " 2 is the attain 
ment by the senses, irrespective of the body, of powers 
directed to objects in any desired place or time ; "the con 
quest of nature " is the power of controlling all the mani 
festations of nature. These perfections appear to the full 
in the third kind of yogin, from the subjugation by him of 
the five senses and their essential conditions. 3 These per 
fections are severally sweet, each one by itself, as even a 
particle of honey is sweet, and therefore the second state 
is called Madhupratikd [i.e., that whose parts are sweet]. 

iii. The -Visoku perfection consists in the supremacy 
over all existences, &c. This is said in the aphorisms, 
" To him who possesses, to the exclusion of all other ideas, 
the discriminative knowledge of the quality of goodness 
and the soul, arises omniscience and the supremacy over 
all existences" [iii. 50]. The "supremacy over all ex 
istences " is the overcoming like a master all entities, as 
these are but the developments of the quality of "good 
ness " in the mind [the other qualities of " passion " and 

1 In p. 178, 1. 2, infra, readjpm- aspati explains it as " videhdndm in- 
vritta for pravritti. Cf. Yoga S., driydndm karanabhdvah." 

iii. 52 in Bhoja s Comm. (50 in 3 Vyasa has karanapanchaJcartipa- 
Vydsa s Comm. ) jaya ; Vachaspati explains rupa by 

2 Read vikaranabhdvah ; Vach- yrahanddi (cf. iii. 47). 


" darkness " being already abolished], and exist only in 
the form of energy and the objects to be energised upon. 1 
The discriminative knowledge of them, as existing in the 
modes " subsided," " emerged," or " not to be named," 2 is 
" omniscience." This is said in the aphorisms [i. 36], " Or 
a luminous immediate cognition, free from sorrow 3 [may 
produce steadiness of mind]." 

iv. The Samskdraseshatd state is also called asamprajndta, 
i.e., " that meditation in which distinct recognition of an 
object is lost;" it is that meditation " without a seed" [i.e., 
without any object] which is able to stop the " afflictions" 
that produce fruits to be afterwards experienced in the 
shape of rank, length of life, and enjoyment ; and this 
meditation belongs to him who, in the cessation of all 
modifications of the internal organ, has reached the highest 
" dispassion." " The other kind of meditation [i.e., that 
in which distinct recognition of an object is lost] is pre 
ceded by that exercise of thought which produces the en 
tire cessation of modifications ; it has nothing left but the 
latent impressions" [of thought after the departure of all ob 
jects] [i.e., samsTcdrasesha, i. 1 8]. Thus this foremost of men, 
being utterly passionless towards everything, finds that the 
seeds of the "afflictions," like burned rice-grains, are bereft 
of the power to germinate, and they are abolished together 
with the internal organ. When these are destroyed, there 
ensues, through the full maturity of his unclouded " discri 
minative knowledge," an absorption of all causes and effects 
into the primal prakriti ; and the soul, which is the power 
of pure intelligence, abiding in its own real nature, and 
escaped from all connection with the phenomenal under 
standing (buddhi), or with existence, reaches " absolute 
isolation" (kaivalyd). Final liberation is described by Patan- 
jali as two perfections : " Absolute isolation is the repressive 
absorption 4 of the qualities which have consummated 

1 I read in p. 179, 1. II, vyava- 3 Visokd. 

sdj/avyavaseydtmaJcdndm, from Vya- 4 This is explained by Vdchaspati, 

sa s Comm. " The latent impressions produced 

* I.e., as past, present, or future. by the states of the internal organ 


the ends of the soul, i.e., enjoyment and liberation, or the 
abiding of the power of intelligence in its own nature " 
[iv. 33]. Nor should any one object, "Why, however, 
should not the individual be born again even though this 
should have been attained ? " for that is settled by the 
well-known principle that "with. the cessation of the 
cause the effect ceases," and therefore this objection is 
utterly irrelevant, as admitting neither inquiry nor de 
cision ; for otherwise, if the effect could arise even in the 
absence of the cause, we should have blind men finding 
jewels, and such like absurdities ; and the popular proverb 
for the impossible would become a possibility. And so, 
too, says the Sruti, " A blind man found a jewel ; one 
without fingers seized it ; one without a neck put it on ; 
and a dumb man praised it." l 

Thus we see that, like the authoritative treatises on 
medicine, the Yoga-sastra consists of four divisions; as 
those on medicine treat of disease, its cause, health, and 
medicine, so the Yoga-sastra also treats of phenomenal 
existence, its cause, liberation, and its cause. This exist 
ence of ours, full of pain, is what is to be escaped from ; 
the connection of nature and the soul is the cause of our 
having to experience this existence ; the absolute abolition 
of this connection is the escape ; and right insight is the 
cause thereof. 2 The same fourfold division is to be similarly 
traced as the case may be in other Sastras also. Thus all 
has been made clear. 

called vyuttJidna (when it is chiefly ment of these qualities when one 

characterised by activity, or dark- or another becomes predominant, 

ness, iii. 9) and nirodha (when it is l This curious, passage occurs in 

chiefly characterised by the quality the Taittiriya - Aranyaka i. 1 1, 5. 

of goodness ), are absorbed in the Mddhava in his Comment, there 

internal organ itself ; this in egoism explains it of the soul, and quotes 

(asmitd) ; egoism in the merely the Svetdsv. Up., iii. 19. Mddhava 

once resolvable (i.e., buddhi) ; and here takes avindat as " he pierced 

buddhi into the irresolvable (i.e., the jewel," but I have followed his 

prakriti)." Prakriti consists of the correct explanation in the Comm. 

three qualities in equilibrium; and 2 This is taken from Vachaspati s 

the entire creation, consisting of Comm. on Yoga S. ii. 15. Cf. the 

causes and effects, is the develop- " four truths " of Buddhism. 


The system of Sankara, which comes next in succession, 
and which is the crest-gem of all systems, has been ex 
plained by us elsewhere ; it is therefore left untouched 
here. 1 E. B. C. 


There is an interesting description of the Yogins on the Mountain 
Eaivataka in Magha (iv. 55) : 

"There the votaries of meditation, well skilled in benevolence 
(maitri) and those other purifiers of the mind, having successfully 
abolished the afflictions ; and obtained the * meditation possessed 
of a seed, and having reached that knowledge which recognises 
the essential difference between the quality Goodness and the Soul, 
desire yet further to repress even this ultimate meditation." 

It is curious to notice that maitri, which plays such a prominent 
part in Buddhism, is counted in the Yoga as only a preliminary 
condition from which the votary is to take, as it were, his first start 
towards his final goal. It is called a parikarman ( = prasddhaka) in 
Vyasa s Comm. i. 33 (cf. iii. 22), whence the term is borrowed by 
Magha. Bhoja expressly says that this purifying process is an 
external one, and not an intimate portion of yoga itself; just as in 
arithmetic the operations of addition, &c., are valuable, not in them 
selves, but as aids in effecting the more important calculations which 
arise subsequently. The Yoga seems directly to allude to Buddhism 
in this marked depreciation of its cardinal virtue. 


For the word vydkopa in the original here (see also p. 242, 1. 3 
infra), cf. Kusumanjali, p. 6, 1. 7. 

1 This probably refers to the Pan- tddhycaya-brahmana, p. x), but, if 

chadasi. A Calcutta Pandit told this is the same as the vivarana- 

ire that it referred to the Prameya- prameya-sangraha, it is by Bha"ra- 

vivarana-sangraha (cf. Dr. Burnell s titirthavidyaranya (see Dr. Burnell s 

preface to his edition of the Deva- Cat of Tanjore MSS. p. 88). 


ON THE UPADHI (cf. supra, pp. 7, 8, 174, 194). 

[As the upddhi or " condition " is a peculiarity of 
Hindu logic which is little known in Europe, I have 
added the following translation of the sections in the 
Bhasha-parichchheda and the Siddhanta-muktavali, which 
treat of it.] 

cxxxvii. That which always accompanies the major term 
(sddhya), ~but does not always accompany the middle 
(Tietu), is called the Condition (upddhi) ; its examina 
tion is now set forth. 

Our author now proceeds to define the upddhi or 
condition, 1 which is used to stop our acquiescence in a 
universal proposition as laid down by another person ; 
" that which always accompanies," &c. The meaning of 
this is that the so-called condition, while it invariably 

1 The upddhi is the " condition " smoke. Similarly, the alleged ar- 
which must be supplied to restrict gument that " B is dark because he 
a too general middle term. If the is Mitra s son " fails, if we can estab- 
middle term, as thus restricted, is lish that the dark colour of her for- 
still found in the minor term, the mer offspring A depended not on 
argument is valid ; if not, it fails, his being her son, but on her hap- 
Thus, in " The mountain has smoke pening to have fed on vegetables 
because it has fire" (which rests on instead of ghee. If we can prove 
the false premiss that " all fire is ac- that she still keeps to her old diet, 
companied by smoke "), we must add of course our amended middle term 
"wet fuel "as the condition of "fire;" will still prove B to be dark, but 
and if the mountain has wet fuel not otherwise, 
as well as fire, of course it will have 


accompanies that which is accepted as the major term, 
does not thus invariably accompany that which our oppo 
nent puts forward as his middle term. [Thus in the false 
argument, " The mountain has smoke because it has fire," 
we may advance " wet fuel," or rather " the being produced 
from wet fuel," as an upddhi, since " wet fuel " is neces 
sarily found wherever smoke is, but not always where fire 
is, as e.g., in a red-hot iron ball.] 

" But," the opponent may suggest, " if this were true, 
would it not follow that (a) in the case of the too wide 
middle term in the argument, This [second] son of Mitra s, 
whom I have not seen, must be dark because he is Mitra s 
son/ we could not allege the being produced from feeding 
on vegetables l as a condition, inasmuch as it does not 
invariably accompany a dark colour, since a dark colour 
does also reside in things like [unbaked] jars, &c., which 
have nothing to do with feeding on vegetables ? (&) 
Again, in the argument, The air must be perceptible to 
sense 2 because it is the site of touch, we could not allege 
the possessing proportionate form as a condition ; be 
cause perceptibility [to the internal sense] is found in the 
soul, &c., and yet soul, &c., have no form [and therefore the 
possessing proportionate form does not invariably accom 
pany perceptibility], (c) Again, in the argument, Destruc 
tion is itself perishable, because it is produced/ we could 
not allege as a condition the being included in some 
positive category of existence 3 [destruction being a 
form of non-existence, called " emergent/ dvam&dbkdva], 

1 The Hindus think that a child s fire, are spartavat, but by si. 27 of 
dark colour comes from the mother s these air is neither pratyaksha nor 
livin^ on vegetables, while its fair rupavat. 

colour comes from her living on a This condition would imply ;that 

i we could only argue from this middle 

2 By Bhdsha-parich. fl. 25, the term "the being produced "in cases of 
four elements, earth, water, air, and positive existence, not non-existence. 


inasmuch as perishability is found in antecedent non- 
existence, and this certainly cannot be said to be included 
in any positive category of existence." 

We, however, deny this, and maintain that the true mean 
ing of the definition is simply this, that whatever fact or 
mark we take to determine definitely, in reference to the 
topic, the major term which our condition is invariably to 
accompany, that same fact or mark must be equally taken 
to determine the middle term which our said condition is 
not invariably to accompany. Thus (a) the " being pro 
duced from feeding on vegetables " invariably accompanies 
" a dark colour," as determined by the fact that it is Mitra s 
son, whose dark colour is discussed [and this very fact is 
the alleged middle term of the argument ; but the pre 
tended contradictory instance of the dark jar is not in 
point, as this was not the topic discussed]. (&) Again, 
" possessing proportionate form " invariably accompanies 
perceptibility as determined by the fact that the thing 
perceived is an external object ; while it does not in 
variably accompany the alleged middle term " the being 
the site of touch," which is equally to be determined by the 
fact that the thing perceived is to be an external object. 1 
(c) Again, in the argument " destruction is perishable 
from its being produced," the "being included in some 
positive category of existence " invariably accompanies 
the major term "perishable," when determined by the 
attribute of being produced. [And this is the middle term 
advanced; and therefore the alleged contradictory in 
stance, " antecedent non-existence," is not in point, since 
nobody pretends that this is produced at all.] 

But it is to be observed that there is nothing of this 
kind in valid middle terms, i.e., there is nothing there 

1 " Soul," of course, is not external ; but our topic was not soul, but air. 


which invariably accompanies the major term when 
determined by a certain fact or mark, and does not so 
accompany the middle term when similarly determined. 
This is peculiar to the so-called condition. [Should the 
reader object that " in each of our previous examples there 
has been given a separate determining mark or attribute 
which was to be found in each of the cases included under 
each ; how then, in the absence of some general rule, 
are we to find out what this determining mark is to be in 
any particular given case ? " We reply that] in the case 
of any middle term which is too general, the required 
general rule consists in the constant presence of one or 
other of the following alternatives, viz., that the subjects 
thus to be included are either (i.) the acknowledged site 
of the major term, and also the site of the condition, 1 or 
else (ii.) the acknowledged site of the too general middle 
term, but excluding the said condition ; 2 and it will be 
when the case is determined by the presence of one or 
other of these alternatives that the condition will be con 
sidered as " always accompanying the major term, and not 
always accompanying the middle term." 3 

1 As, e.g., the mountain and thoxigh possessing the respective 
Mitra s first son in the two false middle terms "fire " and "the being 
arguments, "The mountain has Mitrd s offspring" do not possess the 
smoke because it has fire " (when respective conditions " wet fuel " or 
the fire-possessing red-hot iron ball " the mother s feeding on vege- 
has no smoke), and " Mitrd s first tables," nor, consequently, the 
son A is dark because he is respective major terms (sddhya) 
Mitrd s offspring " (when her second " smoke " and " dark colour." 

son B is fair). These two subjects 3 This will exclude the objected 

possess the respective sddhyas or case of " dark jars " in (a), as it 

major terms " smoke " and " dark falls under neither of these two alter- 

colour," and therefore are respec- natives ; for, though- they are the 

tively the subjects where the con- sites of the sddhya "dark colour," 

ditions " wet fuel " and " the they do not admit the condition 

mother s feeding on vegetables " are " the feeding on vegetables," nor 

to be respectively applied. the middle term " the being 

2 As, e.g., the red-hot ball of iron Mitrd s son." 
and Mitrd s second son ; as these, 


cxxxviii. All true Conditions reside in the same subjects with 
their major terms; l and, their subjects being thus com 
mon, the (erring) middle term will be equally too general 
in regard to the Condition and the major term. 2 

cxxxix. It is in order to prove faulty generality in a 
middle term that the Condition has to be employed. 

The meaning of this is that it is in consequence of the 
middle term being found too general in regard to the 
condition, that we infer that it is too general in regard 
to the major term ; and hence the use of having a con 
dition at all. (a.) Thus, where the condition invariably 
accompanies an unlimited 3 major term, we infer that the 
middle term is too general in regard to the major term, 
from the very fact that it is too general in regard to the 
condition ; as, for example, in the instance " the mountain 
has smoke because it has fire," where we infer that the 
" fire " is too general in regard to " smoke," since it is too 
general in regard to <c wet fuel ; " for there is a rule that 
what is too general for that which invariably accompanies 
must also be too general for that which is invariably 
accompanied. (&.) But where we take some fact or mark 
to determine definitely the major term which the condition 
is invariably to accompany, there it is from the middle 
term s being found too general in regard to the condition in 
cases possessing this fact, or mark that we infer that the 
middle term is equally too general in regard to the major 
term. Thus in the argument, " B is dark because he is 
Mitra s son," the middle term "the fact of being Mitra s 

1 I.e., wherever there is fire pro- ball of iron), there the upddhi also 
duced by wet fuel there is smoke, is not applicable. 

The condition and the major term 3 I.e., one which requires no deter- 

are "equipollent" in their extension, mining fact or mark, such as the 

2 Where the hetu is found and three objected arguments required 
not the sddhya (as in the red-hot in 137. 


son " is too general in regard to the sddhya, " dark colour," 
because it is too general in regard to the upddhi, u feeding 
on vegetables," as seen in the case of Mitra s second son 
[Mitra s parentage being the assumed fact or mark, and 
Mitra herself not having fed on vegetables previous to his 

[But an objector might here interpose, " If your defini 
tion of a condition be correct, surely a pretended condi 
tion which fulfils your definition can always be found 
even in the case of a valid middle term. For instance, in 
the stock argument the mountain must have fire because 
it has smoke/ we may assume as our pretended condition 
the being always found elsewhere than in the moun 
tain; since this certainly does not always accompany 
the middle term, inasmuch as it is not found in the 
mountain itself where the smoke is acknowledged to be ; 
and yet it apparently does always accompany the major 
term/ since in every other known case of fire we certainly 
find it, and as for the present case you must remember 
that the presence of fire in this mountain is the very point 
in dispute." To this we reply] You never may take*such 
a condition as " the being always found elsewhere than in 
the subject or minor term " (unless this can be proved by 
some direct sense-evidence which precludes all dispute) ; 
because, in the first place, you cannot produce any argu 
ment to convince your antagonist that this condition does 
invariably accompany the major term [since he naturally 
maintains that the present case is exactly one in point 
against you] ; and, secondly, because it is self -contradictory 
[as the same nugatory condition may be equally employed 
to overthrow the contrary argument]. 

But if you can establish it by direct sense-evidence, then 
the " being always found elsewhere than in the subject " 



becomes a true condition, [and serves to render nugatory 
the false argument which a disputant tries to establish]. 
Thus in the illusory argument " the fire must be non-hot 
because it is artificial," we can have a valid condition in 
" the being always found elsewhere than in fire," since we 
can prove by sense-evidence that fire is hot, 1 [thus the 
upddhi here is a means of overthrowing the false argu 

Where the fact of its always accompanying the major 
term, &c., is disputed, there we have what is called a 
disputed condition. 2 But "the being found elsewhere 
than in the subject " can never be employed even as a dis 
puted condition, in accordance with the traditional rules 
of logical controversy. 3 

E. B. C. 

1 The disputant says, " Fire must 
be non-hot because it is artificial." 
"Well," you rejoin, "then it must 
only be an artificiality which i^ al 
ways found elsewhere than in fire, 
i.e., one which will not answer 
your purpose in trying to prove 
your point." Here the proposed 
upddhi "the being always found 
elsewhere than in fire " answers to 
the definition, as it does not always 
accompany the hetu " possessing arti 
ficiality," but it does always accom 
pany the sddliya " non-hot," as fire is 
proved by sense-evidence to be hot. 

2 As in the argument, " The earth, 
&c., must have had a maker because 
they have the nature of effects," 
where the Theist disputes the Athe 
istic condition " the being produced 
by one possessing a body." See 
Kusumdnjali, v. 2. 

3 In fact, it would abolish all dis 
putation at the outset, as each 
party would produce a condition 
which from his own point of view 
would reduce his opponent to si 
lence. In other words, a true con 
dition must be consistent with either 
party s opinions. 



University of Toronto 








Acme Library Card Pocket